1. The method for detecting the residual emergency maintenance time of the equipment comprises

Acquiring an alarm strategy;

the early warning point detection is carried out according to the obtained warning strategy, which comprises

Judging whether an early warning value exists in the abnormal values determined currently according to an alarm strategy;

and when the early warning value is judged to exist, determining the latest early warning point according to the early warning value.

2. The method of claim 1, wherein the remaining emergency repair time is a remaining emergency repair time for a liquid helium pressure value of a medical device from a leak threshold at which a leak occurs, the acquisition alarm strategy implemented to include

And acquiring an alarm strategy corresponding to the liquid helium pressure of the current equipment from a pre-configured and stored abnormal configuration information database according to the currently detected equipment ID.

3. The method of claim 1 or 2, wherein the alarm policy is a persistent worsening alarm policy, and wherein determining whether an alarm value exists among currently determined outliers according to the alarm policy is implemented to include

When the abnormal liquid helium pressure value of the equipment is detected, comparing the current abnormal value with the previous early warning value, judging whether the current abnormal value is deteriorated compared with the previous early warning value according to the comparison result, and determining the current abnormal value as the early warning value when the deterioration is judged; the deterioration refers to a trend that the abnormal value is further away from the preset alarm threshold value compared with the previous early warning value;

when the early warning value is judged to exist, the latest early warning point is determined according to the early warning value to be realized as including

And determining the time point of the early warning value as the latest early warning point needing to be warned and reminded.

4. The method of claim 1 or 2, wherein the alarm policy is a persistent worsening alarm policy, and wherein determining whether an alarm value exists among currently determined outliers according to the alarm policy is implemented to include

Acquiring the maximum value of the current day from the abnormal data information record of the liquid helium pressure of the current equipment;

judging whether the obtained maximum value obtained by the current day is an early warning value or not, and determining the maximum value as the early warning value if the judgment result is negative;

when the early warning value is judged to exist, the latest early warning point is determined according to the early warning value to be realized as including

And determining the time point of the early warning value as the latest early warning point needing to be warned and reminded.

5. The method of claim 1 or 2, wherein the alarm policy is a period-limited alarm policy, and wherein determining whether an alarm value exists among currently determined outliers according to the alarm policy is implemented to include

Acquiring an alarm period, an upper limit threshold and a lower limit threshold set for a period limit alarm strategy of the liquid helium pressure from a pre-configured and stored abnormal configuration information database;

performing abnormity detection according to an alarm period, acquiring the latest liquid helium pressure value when the time interval of the alarm period is reached, comparing the latest liquid helium pressure value with a preset upper limit threshold value and a preset lower limit threshold value respectively, and determining the latest liquid helium pressure value as an early warning value when the latest liquid helium pressure value is greater than the upper limit threshold value or less than the lower limit threshold value;

when the early warning value is judged to exist, the latest early warning point is determined according to the early warning value to be realized as including

And determining the occurrence time of the determined early warning value as the latest early warning point needing to be warned.

6. The method of claim 4, wherein the alarm period comprises a first alarm period and a second alarm period, the period duration of the first alarm period being greater than the period duration of the second alarm period;

before timing according to the alarm period, switching between a first alarm period and a second alarm period according to the alarm frequency, so that when the alarm is frequent, the second alarm period is adopted for detecting the early warning value, and when the alarm is sparse, the first alarm period is adopted for detecting the early warning value;

and the alarm frequency is the frequency of carrying out alarm reminding according to the determined latest early warning point.

7. The method of claim 1 or 2, wherein the alarm policy is a fluctuation range alarm policy, and wherein determining whether an early-warning value exists among currently determined outliers according to the alarm policy is implemented to include

Acquiring an alarm period, a fluctuation range, an upper limit threshold and a lower limit threshold which are set for a liquid helium pressure fluctuation range limited alarm strategy from a pre-configured and stored abnormal configuration information database;

timing according to the alarm period, and acquiring the maximum value and the minimum value of the liquid helium pressure in the time interval of the alarm period from the liquid helium pressure information database when the time interval of the alarm period is reached;

calculating the fluctuation value of the liquid helium pressure in the alarm period according to the obtained maximum value and the minimum value;

calculating a fluctuation range threshold value of the liquid helium pressure according to the fluctuation range, the upper limit threshold value and the lower limit threshold value;

determining an early warning value according to the fluctuation value and the fluctuation range threshold value;

when the early warning value is judged to exist, the latest early warning point is determined according to the early warning value to be realized as including

And determining the time point of the determined early warning value as the latest early warning point needing to be warned.

8. The method for early warning the remaining emergency maintenance time of the equipment comprises the following steps:

acquiring a liquid helium pressure value and occurrence time of the liquid helium pressure value in real time, and collecting abnormal data information of the liquid helium pressure according to the acquired real-time liquid helium pressure value and a preset alarm threshold, wherein the alarm threshold comprises an upper threshold and a lower threshold;

determining a latest early warning point by an early warning point detection method of remaining emergency maintenance time of the apparatus according to any one of claims 1 to 7;

and when the latest early warning point is obtained, generating the remaining emergency maintenance time in a curve fitting mode according to the collected abnormal data information record and the preset leakage threshold value, and generating and outputting alarm information according to the generated remaining emergency maintenance time and the obtained latest early warning point.

9. The method of claim 8, wherein generating the remaining emergency repair time by curve fitting based on the collected abnormal data information record and the preset leak threshold comprises:

judging whether the record quantity in the abnormal data information record meets an estimation condition, and acquiring abnormal data within a certain time range from the latest early warning point when the estimation condition is met, wherein the estimation condition is the quantity of input data required by a selected fitting curve;

calculating fitting parameters through curve fitting according to the acquired abnormal data within a certain time range from the latest early warning point;

and generating the residual emergency maintenance time according to the preset leakage threshold value and the calculated fitting parameters.

10. The residual emergency maintenance time early warning system of the equipment comprises an information acquisition module, a system platform and a user terminal, wherein,

the information acquisition module is connected with the equipment and is used for acquiring liquid helium pressure data of the equipment in real time;

the system platform can be respectively communicated with the information acquisition module and the user terminal, and is used for acquiring real-time liquid helium pressure data acquired by the information acquisition module, collecting abnormal data information and performing early warning point analysis according to a preset warning strategy, a preset warning threshold value and the real-time liquid helium pressure data, and generating warning information and outputting the warning information to the user terminal according to the collected abnormal data information and the residual emergency maintenance time of the latest early warning point calculation equipment determined by analysis.

Background

Liquid helium pressure is a key parameter in nuclear magnetic resonance equipment, once the liquid helium pressure fluctuates abnormally, serious consequences such as liquid helium leakage and even quench are likely to occur, and huge loss is caused to hospitals. Therefore, when the liquid helium may leak due to abnormal fluctuation of the liquid helium pressure, related personnel need to be informed to take treatment measures in time, so as to avoid serious consequences and loss.

At present, whether liquid helium leakage is possible to be caused by liquid helium pressure is judged by a maintenance engineer according to the description of hospital repair, and how long the liquid helium leakage occurs in the equipment is estimated according to experience, and then a decision is made on which treatment measure is taken. Therefore, the existing processing modes are manually inquired and judged, the personal subjective judgment of engineers is completely relied on, and the hospital arrangement personnel is required to carry out the guard and the maintenance, so the labor cost is higher, and the processing efficiency is not high. Therefore, there is a need to provide a solution that is more effective, more timely, and more accurate in estimating the remaining time of liquid helium leakage.

Disclosure of Invention

According to one aspect of the invention, the method for estimating the remaining emergency maintenance time of the monitoring parameters of the equipment is provided, so as to solve the problems that in the prior art, the time for the liquid helium pressure abnormal distance leakage needs to be subjectively judged by a maintenance engineer, and the accuracy and the efficiency are low. The method can be extended to other monitoring parameters of other equipment from the liquid helium pressure parameter of the nuclear magnetic resonance equipment, so that the problem that equipment maintenance depends on manual judgment of a maintenance engineer at present is solved, the abnormal condition of the monitoring parameters is automatically analyzed based on the monitoring parameter data of the equipment, and the residual emergency maintenance time of the equipment is estimated based on the abnormal condition.

The method for estimating the remaining emergency maintenance time of the equipment comprises the following steps: configuring monitoring parameters, alarm thresholds and larger hazard threshold storage of the equipment; acquiring a monitoring parameter value and occurrence time of the equipment in real time, and collecting abnormal data according to the real-time monitoring parameter value and an alarm threshold value; generating a remaining emergency repair time based on the collected abnormal data information record and the greater hazard threshold. By the method, the alarm threshold value can be configured according to actual experience and equipment conditions, and the larger damage threshold value is configured according to a manufacturer technical manual, so that the estimation of the residual time of the equipment monitoring parameter reaching the larger damage threshold value is realized, accurate residual emergency maintenance time is provided for a user, the user can timely perform corresponding processing according to a calculation result, the user time is saved, and great loss is avoided.

In some embodiments, generating the remaining emergency repair time based on the collected abnormal data information records and the greater hazard threshold is accomplished by way of curve fitting. Through the curve fitting mode, a plurality of input data can be utilized for fitting, and the obtained result is more accurate.

In some embodiments, collecting anomaly data based on the real-time monitoring parameter values and the alarm thresholds comprises: and comparing the acquired real-time monitoring parameter value with an alarm threshold value, judging whether the real-time monitoring parameter value is an abnormal value tending to the direction change of the larger damage threshold value, and generating abnormal data information for recording and storing according to the abnormal value and the occurrence time thereof when the real-time monitoring parameter value is judged to be the abnormal value tending to the direction change of the larger damage threshold value. When an abnormality occurs, the abnormal data is collected, and the base number of the abnormal data record used as the calculation basis is increased, so that the estimation result is more accurate. And because the collected abnormal data is an abnormal value which tends to change in a larger harm threshold direction, the residual emergency maintenance time is estimated based on the abnormal data record, the reference value is higher, and the prediction result is more accurate.

In some embodiments, generating the remaining emergency repair time based on the collected anomaly data information records and the greater hazard threshold comprises: judging whether the abnormal value is the current extreme value in the abnormal data information record on the same day according to the occurrence time of the abnormal value, and generating the residual emergency maintenance time in a curve fitting mode according to the occurrence time of the abnormal value, the abnormal data information record and the larger hazard threshold value when the abnormal value is the extreme value; or generating the residual emergency maintenance time in a curve fitting mode directly according to the occurrence time of the abnormal value, the abnormal data information record and the larger damage threshold value. The estimation of the remaining emergency maintenance time can be performed directly based on the abnormal data record when the abnormality occurs, or can be performed under the condition that the abnormality is continuously worsened (namely under the condition that the current value is an extreme value, for a parameter which exceeds an upper limit threshold and reaches a larger harm threshold in an ascending trend development mode, the extreme value corresponds to the maximum value, for a parameter which is smaller than a lower limit threshold and reaches the larger harm threshold in a descending trend development mode, the extreme value corresponds to the minimum value), the estimation is performed based on the abnormal data record, when the estimation is performed under the condition of continuous deterioration, the further optimization of the estimation method can be realized, namely, under the condition that the collected real-time data is denser, the repeated calculation of some slightly fluctuating abnormal values is performed, the reference value of the estimation result is improved, and the noise is reduced.

In some embodiments, the curve fitting is a linear fitting, and generating the remaining emergency repair time by the curve fitting according to the occurrence time of the outlier, the outlier data information record, and the greater hazard threshold comprises: judging whether the number of records in the abnormal data information record, the occurrence time of which is the same day as the abnormal value, is not less than two, and if not, calculating the value of the linear fitting parameter by a least square method according to the abnormal value and the occurrence time thereof recorded in the abnormal data information record; and obtaining a larger damage threshold value, and calculating and generating the residual emergency maintenance time through a linear fitting formula according to the linear fitting parameters and the larger damage threshold value. The fitting mode is carried out by the least square method, only two input data are needed as a basis, the estimation can be realized, the realization is simple, the result is more accurate when the estimation of the current remaining time is carried out along with the increase of the data, and the applicable data width is wider.

In some embodiments, calculating the values of the linear fitting parameters by a least squares method based on the outliers recorded in the outlier data information record and their times of occurrence comprises: acquiring an information record serving as input data from the abnormal data information record, taking an abnormal value in the information record serving as the input data as a y value and taking the occurrence time as an x value, and calculating values of a slope v and a constant b by a least square normal fitting formula y ═ v × x + b; calculating and generating the remaining emergency repair time according to the linear fitting parameters and the greater hazard threshold comprises: according to the obtained values of the slope v and the constant b, the larger hazard threshold value is used as a value y, and the occurrence time of the equipment reaching the larger hazard threshold value y is calculated through a least square normal fitting formula y-v x + b; and acquiring the occurrence time of the current abnormal value, and calculating the residual emergency maintenance time according to the occurrence time of the current abnormal value and the calculated occurrence time of the larger damage threshold value y. The current change rate of the monitoring parameters is obtained through least square normal fitting, the change rate is estimated based on the change rate, and the calculated residual emergency maintenance time is predicted based on the current real-time rate due to the fact that the change rate is dynamically calculated based on real-time data, the result is accurate, the method is easy and convenient to achieve, and therefore the prediction method is fast and efficient.

According to another aspect of the present invention, there is also provided a remaining emergency repair time warning method for a device, including: configuring monitoring parameters, an alarm threshold, a larger harm threshold and early warning strategy information storage of equipment; acquiring a monitoring parameter value and occurrence time of the monitoring parameter value in real time, and collecting abnormal data information of the monitoring parameter according to the acquired real-time monitoring parameter value and an alarm threshold value for storage; acquiring early warning strategy information of the monitoring parameters, and acquiring the latest early warning points according to the real-time monitoring parameter values, the warning threshold values, the early warning strategy information and the collected abnormal data information; and when the latest early warning point is obtained, generating the residual emergency maintenance time according to the collected abnormal data information record and a preset larger damage threshold, and generating alarm information according to the generated residual emergency maintenance time and the obtained latest early warning point and outputting the alarm information. Early warning point detection is carried out by setting an early warning strategy, early warning points meeting conditions can be detected according to requirements, when the latest early warning point is detected, the residual emergency maintenance time is estimated, and finally, the early warning point information and the residual emergency maintenance time are jointly used as warning messages to be output to a user. In addition, in the mode, an early warning strategy can be formulated according to requirements, message reminding is carried out based on the early warning strategy, message noise is reduced, really useful important abnormal messages are highlighted, and user experience is greatly improved. In addition, the residual emergency maintenance time is estimated based on the early warning point, so that the system is prevented from frequently calculating the residual emergency maintenance time, and the resource occupation is reduced.

In some embodiments, generating the remaining emergency repair time based on the collected abnormal data information record and the preset greater hazard threshold is implemented by curve fitting, and includes: judging whether records in the abnormal data information records meet estimation conditions or not, and acquiring abnormal data within a certain time range from the latest early warning point when the records meet the estimation conditions; calculating fitting parameters through curve fitting according to the acquired abnormal data within a certain time range from the latest early warning point; and generating the residual emergency maintenance time according to the preset larger damage threshold value and the calculated fitting parameters. The method has the advantages that the residual emergency maintenance time is estimated in the early warning point in a curve fitting mode, the method is simple to realize, the accuracy of the result can be guaranteed by selecting the input data volume, the estimation can be carried out timely, the resource occupation is reduced, and the resource waste is avoided.

In some embodiments, the early warning policy information includes an alarm policy, an alarm period, and a fluctuation range, the alarm policy being a continuous worsening alarm policy, a period limiting alarm policy, or a fluctuation range limiting alarm policy. The continuous deterioration alarm strategy can realize alarm when data is continuously deteriorated, highlight important messages and ensure real-time performance; noise interference can be reduced through a period limited alarm strategy, and the method is suitable for key parameters with small data change frequency; the fluctuation range limiting alarm strategy is optimization of a continuous deterioration alarm strategy, and is not only used for judging the maximum value at a certain moment, but also used for judging abnormality based on the fluctuation range, so that the abnormal condition of data can be more accurately reflected.

In some embodiments, the curve fitting manner is a linear fitting, and when the curve fitting manner is a linear fitting, the fitting parameters are a slope v and a constant b, and the fitting parameters are calculated by a least square method. Therefore, the estimation is carried out through linear fitting, the effect that the estimation result can be obtained based on two input data can be achieved, the effect that the accuracy of the estimation result is improved based on a plurality of input data can be achieved, the adaptability is stronger, and the application range is wider. In other implementation manners, the curve fitting manner can be exponential fitting, logarithmic fitting or trigonometric function fitting according to specific conditions and requirements, and estimation is performed based on other nonlinear fitting manners, so that the requirements of different occasions can be met, and the accuracy of estimation results under different situations can be improved.

According to another aspect of the invention, a system for estimating remaining emergency repair time of equipment is also provided, which comprises an information configuration module, a data storage module and a data processing module, wherein the information configuration module is used for configuring monitoring parameters, alarm thresholds and larger hazard threshold storage of the equipment; the abnormal data collection module is configured to acquire the monitoring parameter value and the occurrence time of the monitoring parameter value of the equipment in real time, and collect and store abnormal data information according to the acquired real-time monitoring parameter value and a preset alarm threshold value; and the residual emergency maintenance time estimation module is configured to generate residual emergency maintenance time output in a curve fitting mode according to the collected abnormal data information record and a preset larger damage threshold value. By the aid of the estimation system, automatic processing of data acquisition and anomaly analysis can be achieved, automatic prediction of emergency maintenance time can be achieved, manpower dependence is liberated, labor cost is reduced, more experience data can be collected quickly, and results are accurate. Moreover, the estimation system can realize data acquisition and analysis estimation of a plurality of devices at the same time, and has very high efficiency.

According to another aspect of the invention, the invention also provides a residual emergency maintenance time early warning system of equipment, which comprises an information acquisition module, a system platform and a user terminal, wherein the information acquisition module is connected with the equipment and is used for acquiring monitoring parameter data of the equipment in real time; the system platform can be respectively communicated with the information acquisition module and the user terminal, and is used for acquiring real-time monitoring parameter data acquired by the information acquisition module, collecting abnormal data information and analyzing early warning points according to a preset alarm strategy, an alarm threshold value and the real-time monitoring parameter data, and generating alarm information and outputting the alarm information to the user terminal according to the collected abnormal data information and the analyzed residual emergency maintenance time of the early warning point computing equipment. Through the early warning system, automatic monitoring and abnormal early warning of equipment monitoring parameters are achieved, the time when the distance between the monitoring parameters reaches greater harm can be accurately estimated, and accurate time reference data are provided for users. And the early warning system can also inform the user terminal bound by the user of the alarm information and the residual time, so that the user can check the alarm information and the residual time very conveniently. Moreover, the system can realize the monitoring and management of one user on a plurality of devices, thereby greatly improving the working efficiency of the user.

In some embodiments, the system platform comprises a parameter setting module, an early warning point acquisition module, a remaining emergency maintenance time estimation module and an alarm module, wherein the parameter setting module is used for configuring monitoring parameters, an alarm threshold, a larger hazard threshold and early warning strategy information storage for the equipment; the early warning point acquisition module is used for acquiring the latest early warning point according to the real-time monitoring parameter data, the alarm threshold value, the larger harm threshold value and the early warning strategy information; the residual emergency maintenance time estimation module comprises an abnormal data collection unit and a time calculation unit, wherein the abnormal data collection unit is used for generating and storing abnormal data information records according to real-time monitoring parameter data and a preset alarm threshold value, and the time calculation unit is used for generating residual emergency maintenance time in a curve fitting mode according to the latest early warning point information, the generated abnormal data information records and a preset larger hazard threshold value when the early warning point acquisition module acquires the latest early warning point; and the alarm module is used for generating alarm information according to the generated residual emergency maintenance time and the obtained latest early warning point and outputting the alarm information to the user terminal. By acquiring the real-time monitoring parameter information acquired by the information acquisition module, the early warning analysis and the estimation of the remaining emergency maintenance time are performed based on the alarm threshold, the real-time monitoring parameter data, the larger hazard threshold and the early warning strategy, so that the result is more accurate. The filtering analysis of the warning strategy to the warning points ensures that the warning messages received by the user terminal are important and valuable, and avoids the interference of message noise. And through the automatic acquisition and analysis of the real-time monitoring parameter data of the equipment, the monitoring and analysis processing of any parameter or module of the equipment can be transplanted to a remote system platform, so that the simultaneous monitoring and analysis of a plurality of pieces of equipment based on one system platform are realized, the efficiency is greatly improved, and the monitoring and maintenance cost of the equipment is reduced.

In some embodiments, the user terminal is one or a combination of two or more of WeChat, SMS, email, and client APP. The message early warning is carried out based on the user terminal commonly used by the user, so that the user can conveniently check the message, various selection ways are provided, the user experience can be greatly provided, and the user can conveniently carry out equipment management and maintenance monitoring.

Drawings

FIG. 1 is a schematic flow chart illustrating a method for estimating remaining emergency repair time according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart illustrating a method for generating remaining emergency repair time in the method of FIG. 1;

FIG. 3 is a flow chart illustrating a remaining emergency repair time warning method according to an embodiment of the present invention;

fig. 4 is a schematic flow chart of a method for binding a user terminal as an output terminal in the early warning method shown in fig. 3;

FIG. 5 is a block diagram of a remaining emergency repair time estimation system according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a frame structure of a remaining emergency repair time warning system according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of a remaining emergency repair time warning system according to another embodiment of the present invention;

fig. 8 is a schematic structural diagram of a frame of a remaining emergency repair time warning system according to another embodiment of the present invention.

Detailed Description

The embodiment of the invention provides a method and a system for monitoring parameters of equipment, estimating the remaining emergency maintenance time according to the data of the monitoring parameters and early warning the abnormity of the monitoring parameters and the remaining emergency maintenance time, which are mainly realized by setting an alarm threshold and a larger harm threshold for the monitoring parameters. The alarm threshold value is a threshold value which deviates from a normal value and is abnormal, and the larger damage threshold value is a threshold value which is more serious damage when the value of the monitoring parameter reaches the value, wherein the alarm threshold value needs to be determined by a maintenance engineer according to the actual condition of the equipment by combining with actual experience, and the larger damage threshold value is given by an equipment manufacturer in a use manual. The invention aims to set an alarm threshold and a larger harm threshold for the monitoring parameters of the equipment, carry out abnormity analysis and estimation of the residual emergency maintenance time through the actual data condition of the monitoring parameters of the equipment, and send the estimated residual emergency maintenance time to a user for early warning, so that the user can conveniently carry out corresponding treatment as soon as possible by combining the residual emergency maintenance time, and the larger harm is avoided. The device in the embodiment of the present invention may be any device capable of providing an alarm threshold and a larger hazard threshold for a monitoring parameter, such as a medical device, specifically, for example, a nuclear magnetic resonance device, CT, ultrasound, and the like, and the monitoring parameter may be any device parameter capable of setting the alarm threshold and the larger hazard threshold, such as liquid helium pressure, cold head temperature, shield temperature, water temperature, and water flow. The embodiments of the present invention will be described in detail with an example of monitoring the liquid helium pressure of the nuclear magnetic resonance apparatus. The technical personnel in the field can understand that when the equipment and the monitoring parameters change, only the set alarm threshold value and the larger damage threshold value need to be changed, and the basic concept of the specific estimation method and the early warning method is not changed, so that when the equipment and the monitoring parameters change, the following text can be referred to for adaptive modification, and the abnormity early warning analysis and the estimation of the residual emergency maintenance time of the monitoring parameters of different equipment can be realized.

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Fig. 1 schematically shows a flow of a method for estimating remaining emergency maintenance time of a device according to an embodiment of the present invention, taking the device as a nuclear magnetic resonance device and the monitored parameter as liquid helium pressure as an example, in an embodiment of the present invention, the alarm threshold is a liquid helium pressure alarm threshold, and the greater hazard threshold is a leakage threshold of the liquid helium pressure (i.e., above which liquid helium will leak and cause a more serious hazard), as shown in fig. 1, the method includes:

step S101: and setting and storing a liquid helium pressure alarm threshold value and a leakage threshold value of the equipment.

Receiving the settings of the liquid helium pressure alarm threshold and the leakage threshold of the user through a page, storing the liquid helium pressure alarm threshold and the leakage threshold set for each device into a database according to the settings of the user, namely generating a data table structure comprising a device ID, the liquid helium pressure alarm threshold and the leakage threshold, and storing the received thresholds as the contents of corresponding fields.

The liquid helium pressure alarm threshold is a threshold for starting early warning, the value is set by a field engineer according to experience according to a liquid helium configuration condition of equipment, and includes an upper limit threshold and a lower limit threshold, the upper limit threshold is used for representing an upper limit value of a liquid helium pressure value, the lower limit threshold is used for representing a lower limit value of the liquid helium pressure value, when an actually acquired liquid helium pressure value exceeds the upper limit threshold or is lower than the lower limit threshold, the liquid helium pressure is indicated to be abnormal, for example, the liquid helium pressure corresponding to turning on and off of a heater is configured to be 3.9 and 4.1, according to the configuration condition, the engineer may set the lower limit threshold to be 3.85, and the upper limit threshold to be 4.15. And the leakage threshold is used for indicating that when the liquid helium pressure reaches the value, liquid helium leakage occurs, and is given by a manufacturer magnet manual, and if the leakage pressure of GE nuclear magnetism is 5.25, the leakage threshold of the equipment is set to be 5.25.

Step S102: and acquiring the liquid helium pressure value of the equipment and the occurrence time of the liquid helium pressure value in real time.

The step can be realized in various ways, for example, by arranging a liquid helium pressure sensor on the equipment, and acquiring the liquid helium pressure value and the occurrence time of the equipment in a way of directly reading data of the liquid helium pressure sensor; for another example, the displayed liquid helium pressure value and the occurrence time are read from the equipment in real time, and a special person carries out real-time recording according to the reading result. In a preferred embodiment, the implementation manner of acquiring the liquid helium pressure value of the equipment and the occurrence time thereof in real time is as follows: and acquiring an equipment log in real time, and analyzing the real-time liquid helium pressure value and the occurrence time of each equipment from the equipment log. The method for acquiring data through the device log can be specifically realized as follows: the nuclear magnetic resonance equipment is connected with the data acquisition module, so that the equipment log is read from the equipment through the data acquisition module and uploaded (for example, sent through a network) to the server for storage. The data acquisition module can be set as a data acquisition box, each device is provided with the data acquisition box, the device logs in the device are read through the data acquisition box, and then the read device logs are sent to the cloud server through a network communication module (such as a 3G/4G module) of the data acquisition box to be stored. And then, traversing the equipment log transmitted by the acquisition module for analysis, mainly extracting key information in the equipment log through a feature matching algorithm, wherein the content of the key information is defined according to specific equipment. After the key information is extracted, the following step S103 may be performed as it is, or the extracted information may be stored and the following step S103 may be performed based on the stored information. The operation of step S103 is intuitive after the extraction, that is, the extracted liquid helium pressure value is output to step S103 for comparison, so that the description is not repeated here. The mode of storing after extraction specifically comprises the following steps: and storing the extracted key information in a database as an index unit to generate a liquid helium pressure information database of each nuclear magnetic resonance device. The concrete implementation is as follows: the method comprises the steps of reading an equipment log uploaded by an acquisition module from a cloud server, forming a database structure with key information as an index unit according to preset key information, analyzing the equipment log through a feature matching algorithm, obtaining historical records related to the key information from the equipment log, wherein the historical records comprise liquid helium pressure values and time stamps, and then generating a liquid helium pressure data record of a certain equipment based on an equipment ID of the equipment and storing the liquid helium pressure data record into a database. Thus, a table structure of the magnetic resonance apparatus is formed with the apparatus ID, the liquid helium pressure value, and the occurrence time as fields, and the contents of the corresponding fields extracted from the apparatus log are stored in the table structure, so that a liquid helium pressure information database of the magnetic resonance apparatus is formed, for example, the contents of the database shown in the following table:

device ID

Time of occurrence

Liquid helium pressure value

TABLE 1

Step S103: and judging whether the liquid helium is abnormal or not according to the real-time liquid helium pressure value and a preset liquid helium pressure alarm threshold value, and generating an abnormal data information record according to a judgment result.

When the real-time liquid helium pressure value is analyzed or obtained from the liquid helium pressure information database, the stored preset liquid helium pressure alarm threshold value is obtained, the real-time liquid helium pressure value is compared with the liquid helium pressure alarm threshold value, whether the real-time liquid helium pressure value is larger than the upper limit threshold value or smaller than the lower limit threshold value or not is judged, and if the real-time liquid helium pressure value is larger than the upper limit threshold value or smaller than the lower limit threshold value, the current liquid helium pressure value is abnormal.

When the judgment result is abnormal, an abnormal data information record is generated according to the judgment result. It should be noted that, in the embodiment of the present invention, the monitoring parameter is the liquid helium pressure, and for the liquid helium pressure of the nuclear magnetic resonance apparatus, only the increase of the liquid helium pressure may cause a risk of leakage, so that only the estimation of the remaining maintenance time based on the increased abnormal data of the liquid helium pressure is reasonable and more accurate, and therefore, when the abnormal data information record is generated according to the determination result in the embodiment of the present invention, only the liquid helium pressure value exceeding the upper limit threshold is collected, that is, only the abnormal data higher than the upper limit threshold is generated as the abnormal data information record and stored in the database. In other embodiments, for the generation manner and characteristics of the greater risk threshold of the monitoring parameter, all abnormal real-time data may be recorded to generate an abnormal data information record, or only the real-time data lower than the lower threshold may be recorded to generate an abnormal data information record, the embodiment of the present invention is not considered as a limitation on the characteristics of the collected abnormal data, and the basic concept is as follows: only abnormal data that tends toward a greater hazard threshold (i.e., abnormal data that changes in the direction of the greater hazard threshold) is collected, and an abnormal data information record is generated, such that an estimate of remaining emergency repair time is made based on the abnormal data information record.

Step S104: and generating the residual emergency maintenance time according to the generated abnormal data information record and a preset leakage threshold value.

After a certain amount of abnormal data is collected, the time for the liquid helium pressure to deteriorate to the leakage threshold value can be predicted through a curve fitting mode, and therefore the remaining emergency maintenance time is obtained. The calculation of the remaining emergency repair time by curve fitting may be based on an abnormal condition, that is, after an abnormality exceeding the upper threshold is detected and stored in step S103, the calculation of the remaining emergency repair time may be performed based on the abnormal data record. In the preferred embodiment, the calculation of the remaining emergency repair time may be performed only when the abnormal data is continuously deteriorated. In the embodiment in which the remaining emergency repair time is calculated when an abnormality occurs, the calculation operation for the remaining emergency repair time may be very frequent, and since some abnormalities are not particularly serious, the resource occupancy rate of frequent calculation is high, and the user experience is not good enough. Based on this, in a preferred embodiment, the calculation of the remaining emergency maintenance time is performed only when the abnormal data is deteriorated, wherein the deterioration means that the abnormal condition is more and more serious, that is, the current abnormal liquid helium pressure value is developed towards a direction of being more and more far away from the liquid helium pressure alarm threshold value, and is further away from the liquid helium pressure alarm threshold value than the last abnormality, therefore, after judging whether the abnormality is abnormal or not, whether the abnormality is deteriorated or not is further judged, if the abnormality is continuously deteriorated, the calculation of the remaining emergency maintenance time is performed, otherwise, the calculation is not performed, and only the abnormal data is stored. When the determination result indicates that an abnormality occurs, the method of further determining whether the abnormal value is further deterioration data may be: the method comprises the steps of acquiring a current abnormal value and occurrence time thereof, judging whether the current abnormal value is the maximum value of the current abnormal value in the abnormal data information record in the current day (namely, the current abnormal value is the same day as the occurrence time of the current abnormal value) by a database retrieval mode, if the current abnormal value is the maximum value, indicating that the abnormality is deteriorated compared with the previous abnormality, and if the current abnormal value is not the maximum value, indicating that the abnormality is not deteriorated. And when the deterioration is judged to occur, calculating the residual emergency maintenance time, wherein the specific calculation mode is to estimate based on the abnormal data information record and the leakage threshold value in a curve fitting mode. The curve fitting may be linear fitting or non-linear fitting, such as exponential fitting, logarithmic fitting, trigonometric function fitting, and the like, and the fitting parameters may be calculated by a least square method when linear fitting is performed, and may be calculated by a common fitting parameter implementation algorithm according to the characteristics of the corresponding fitting manner and the prior art when non-linear fitting is performed. Because curve fitting requires a certain amount of input data to obtain output, the number of collected abnormal data information records needs to satisfy a certain amount to meet the condition capable of calculation (i.e., to meet the estimation condition), wherein the collected certain amount is determined according to the selected fitting curve. The embodiment of the invention is described by taking the linear fitting as an example for curve fitting, and other fitting modes can be realized by combining the following implementation modes such as a determination mode of input data and proper adjustment of parameters and formulas in the corresponding fitting modes. In the case of linear fitting, since the number of input data required for linear fitting is at least two, the number of the collected abnormal data information records is not less than two. FIG. 2 illustrates a method of generating remaining emergency repair time in a curve fitting manner for a least squares linear fit, as shown in FIG. 2, the method comprising:

step S1041: it is determined whether the number of records in the abnormal data information record is not less than two, and if not, step S1042 is performed.

The least square normal fitting method is to input 2-n x values and y values and output a slope and a constant, so that whether the data quantity in the abnormal data information record meets the minimum input requirement or not needs to be judged at first, and when the data quantity is not less than two, the calculation operation is carried out, otherwise, under the abnormality of continuous deterioration, the calculation of the residual emergency maintenance time is not carried out. And when the data quantity in the abnormal data information record meets the condition, performing step S1052, and when the condition is not met, continuing data acquisition and processing for judging whether the real-time liquid helium pressure value is abnormal or not.

In a preferred embodiment, the number of the judged abnormal data information records is the number of records on the same day as the occurrence time of the current abnormal value, that is, the number of abnormal records on the same day is not less than two.

Step S1042: from the outlier data records, the slope and constants were calculated by least squares linear fit.

When the number of the abnormal data records is not less than two, the occurrence time is taken as x-axis data, the liquid helium pressure value in the abnormal data record is taken as y-value data, the occurrence time is taken as y-value data, the liquid helium pressure value in the abnormal data record is taken as x-value data, the occurrence time in the abnormal data record is taken as y-value data, the y-value data is taken as x-value data, and the slope v and the constant b in the y-v-x + b are obtained through linear fitting. In this linear fit curve, the slope v represents the rate of increase in the continuing deterioration of the liquid helium pressure value.

Step S1043: and calculating the occurrence time reaching the leakage threshold according to the slope, the constant, the preset leakage threshold and a least square normal linear formula.

And acquiring a preset leakage threshold, and taking the leakage threshold as a y value according to the obtained slope v and the constant b to obtain the occurrence time x of the leakage threshold, namely calculating the occurrence time when the liquid helium pressure value reaches the leakage threshold.

Step S1044: and obtaining the current time, and calculating the remaining emergency maintenance time according to the occurrence time of the leakage threshold and the current time.

The current time of the system at the location of the equipment is obtained (the time of the location of the equipment is used as the standard because the monitored equipment and the cloud server are possibly in different time zones), then the current time is subtracted from the occurrence time of the leakage threshold value, and the remaining emergency maintenance time can be obtained, namely the time when the liquid helium pressure estimated according to the current trend reaches the leakage threshold value can be obtained, and the time is the remaining emergency maintenance time provided for maintenance personnel. In other embodiments, the current time may not be the current time of the system, but may be the occurrence time of the current outlier.

In other embodiments, the method for obtaining the remaining emergency maintenance time through the least square normal linear fitting may further include obtaining a leakage threshold and a current real-time liquid helium pressure value after obtaining the slope v, and obtaining the remaining emergency maintenance time through a formula (leakage threshold-current real-time liquid helium pressure value)/v since the slope v is an estimated liquid helium pressure increase rate.

In a preferred embodiment, the y value and the x value input in step S1042 may not be all the abnormal information in the abnormal data record, but in the case of continuous deterioration, a certain amount of abnormal information may be obtained from the abnormal data record, for example, an abnormal point near (e.g., within ten minutes or twenty minutes) the current deteriorated abnormal value (i.e., the last point to be warned) is extracted as an input (i.e., a liquid helium pressure value of an abnormality within ten minutes or twenty minutes is input as the y value, and an occurrence time is input as the x value), so as to calculate the slope and the constant. In this case, since the selected input point is an abnormal value near the current point to be warned (i.e., the current deteriorated abnormal value), the obtained deterioration rate v is closer to the actual situation, so that the calculated remaining emergency maintenance time is more accurate. The specific implementation of the preferred embodiment is illustrated as follows: assuming that the upper limit of the liquid helium pressure threshold of a certain nuclear magnetic equipment is 6 o' clock in 29/10 in 4.15,2017 years, recording an abnormal value 4.16 in an abnormal data information record, and assuming that the abnormal value does not appear before the day, when judging whether the current abnormal value is continuously deteriorated or not, the abnormal value meets the condition of cutting off the current maximum value on the day, so that the abnormal value is judged to be a continuously deteriorated value, namely an early warning value, at the moment, calculating the remaining emergency maintenance time, obtaining the record number from the abnormal data information record for judgment, and because only one abnormal record exists on the day and the estimation condition is not met, not calculating the remaining time. Then, assuming that point 01 is 6, a second abnormal value 4.19 is recorded in the abnormal data information record, at this time, whether the second abnormal value is continuously deteriorated is judged, since the point 4.19 is the maximum value of the current day cut off, the second abnormal value is judged to be continuously deteriorated, namely, the second abnormal value is an early warning value, at this time, the calculation of the remaining emergency maintenance time is carried out, the number of records is obtained from the abnormal data information record for judgment, since two abnormal records exist in the day, the abnormal value and the occurrence time of the two records are obtained, the two occurrence time points are converted into unix time stamps, namely 1509228000 and 1509228060, the corresponding liquid helium pressure values are respectively 4.16 and 4.19, v is obtained to be 0.0005, b is-7.546 e +05 through linear fitting, x is obtained to be 1509230180 by substituting the liquid helium threshold value 5.25 into the formula, and the UTC time is changed to be 6 points 36 on day 29 of month 29. Since the occurrence time of the abnormal value, i.e., the warning value, is 10 months, 29 days, 6 o' clock, 01 minutes, the remaining emergency repair time to reach the leakage threshold value is 35 minutes at the abnormal value. Next, assuming that a third abnormal value of 4.18 is recorded in the abnormal data information record at point 6 and point 02, the third abnormal value is not the maximum value of the current time of the day, and is therefore data that is not continuously deteriorated, that is, a non-warning value, and therefore, although an abnormality occurs at this time, the remaining emergency maintenance time is not estimated. Assuming that the fourth abnormal value is recorded as 4.21 in the data abnormal information record at point 03 of 6, judging whether the abnormal value is a continuously deteriorated data value, judging that the abnormal value is an early warning value because the current abnormal value which is the fourth abnormal value is the maximum value which is cut off on the same day, calculating the remaining emergency maintenance time, acquiring the four abnormal values (namely, the abnormal points within 10 minutes from the current early warning value which is the last early warning value) and carrying out linear fitting to obtain v which is 0.00023, b which is-3.521 e +05 and the remaining emergency maintenance time which is 1 hour and 14 minutes.

Through the steps, the time reaching the leakage threshold value and the residual emergency maintenance time at the deterioration rate can be estimated according to the deterioration condition of the liquid helium pressure value of the current equipment, analysis and calculation are carried out based on the collected abnormal data of each equipment, the obtained result is more in line with the actual condition, and the reference value is higher and more accurate. Moreover, by the automatic estimation mode, the dependence on the individual subjective judgment capability of a maintenance engineer is liberated, the defect of easy error is avoided, and the informatization of equipment maintenance detection is realized. The skilled person can understand that the above estimation method can also be applied to other monitoring parameters of other equipment, and the remaining emergency maintenance time of the monitoring parameters of other equipment can be obtained by only changing the liquid helium pressure alarm threshold value into the alarm threshold value of the corresponding monitoring parameter, changing the leakage threshold value into the larger damage threshold value of the corresponding monitoring parameter, and performing information acquisition, abnormality judgment and curve fitting based on abnormal data based on the same concept. In the preferred embodiment, the calculation of the remaining emergency repair time may be performed when the anomaly is continuously deteriorated data, that is, when the early warning value is reached, and the input data of the curve fitting may be an anomaly point near the early warning value.

Fig. 3 schematically shows a method flow of a remaining emergency repair time warning method for equipment according to an embodiment of the present invention, and as shown in fig. 3, taking the equipment as a nuclear magnetic resonance equipment and taking a monitoring parameter as liquid helium pressure as an example, the method includes:

step S301: and setting and storing a liquid helium pressure alarm threshold value, a leakage threshold value and an alarm strategy of the equipment.

Presetting a liquid helium pressure alarm threshold, a leakage threshold and an alarm strategy of the equipment, and storing the preset liquid helium pressure alarm threshold, the preset leakage threshold and the preset alarm strategy into the abnormal configuration information database record of the corresponding equipment. And receiving the settings of the liquid helium pressure alarm threshold, the leakage threshold and the alarm strategy by the user through a user page, wherein the setting mode is the same as the method in the step S101, and therefore, the details are not repeated. The setting of the alarm policy may be input or selected through a user page, and the embodiment of the present invention is described by taking selection as an example. In the embodiment of the present invention, the selectable alarm policy configuration may include a continuous deterioration alarm policy, a period limitation alarm policy, and a fluctuation range limitation alarm policy, and the generated abnormal configuration information includes a device ID, an alarm threshold, a leakage threshold, an alarm policy, an alarm period, and a fluctuation range. In other embodiments, other alarm strategies may be set based on actual requirements according to the characteristics and actual conditions of the devices and the key parameters, which are not limited in the embodiments of the present invention, and any modifications and improvements that are made based on the concept of setting a certain optimization strategy to filter alarm points so as to reduce the noise of the alarm message are considered to be within the scope of the present invention. The embodiment of the invention is mainly elaborated based on a continuous deterioration alarm strategy, a period limit alarm strategy and a fluctuation range limit alarm strategy.

The continuous deterioration alarm strategy refers to that when an abnormal value is deteriorated (i.e. the abnormal value is further away from an alarm threshold value), an alarm is given. The alarm policy may be implemented, for example, as: storing the current alarm value after each alarm, when the abnormality is detected again, firstly comparing the current abnormal value with the previous early-warning value to judge whether the abnormality is caused according to the comparison result, if the abnormality is caused, taking the current abnormal value as the early-warning value, namely, if the abnormal value is larger than the upper limit threshold value and the abnormal value is higher than the previous early-warning value, indicating that the abnormal condition is caused to be deteriorated, and then alarming; if the abnormal value is smaller than the lower limit threshold value and the abnormal value is lower than the early warning value of the last time, the abnormal condition is worsened, and an alarm is given at the moment; and if not, no alarm is given.

The period limit alarm strategy refers to the step of designating an alarm period, carrying out one-time abnormal detection in each period, and alarming if the liquid helium pressure value (namely the monitoring parameter value) during detection exceeds an alarm threshold value. The alarm policy may be implemented, for example, as: setting an alarm period for a period limit alarm strategy, timing according to the alarm period, obtaining the latest liquid helium pressure value from a storage unit when the time interval of the alarm period is reached, comparing the liquid helium pressure value with an upper limit threshold value and a lower limit threshold value respectively, if the liquid helium pressure value is larger than the upper limit threshold value or smaller than the lower limit threshold value, taking the value as an early warning value, and taking the occurrence time of the value as an early warning point to alarm. Taking the liquid helium pressure of the nmr as an example, the specific implementation of the strategy may be, for example: setting a period limit alarm strategy for key parameters of liquid helium pressure of certain nuclear magnetic resonance equipment, setting an alarm period to be 30 minutes, obtaining the alarm strategy for judgment, obtaining the alarm period for timing when the period limit alarm strategy is judged, reading a liquid helium pressure value from a storage unit every 30 minutes, obtaining the latest primary liquid helium pressure data value from the primary liquid helium pressure data value, comparing the value with an upper limit threshold value and a lower limit threshold value corresponding to the set liquid helium pressure respectively, and taking the value point as an early warning point for warning when the value exceeds the threshold value. In a preferred embodiment, the period-limited alarm strategy may also be implemented to detect anomalies based on varying alarm periods, such as setting two alarm periods, a first alarm period set to a longer time interval, e.g. 30 minutes, and a second alarm period set to a shorter time interval, e.g. 5 minutes, the anomaly detection of the key parameter is switched between a first alarm period and a second alarm period, namely, a period of time is based on the time interval defined by the first alarm period for carrying out abnormity detection, a period of time is based on the time interval defined by the second alarm period for carrying out abnormity detection, the switching method may be based on the alarm frequency, and when the alarm frequency is higher, the anomaly detection is performed through a shorter alarm period, and when the alarm frequency is lower, the anomaly detection is performed through a longer alarm period. For example, the method can be implemented such that for a certain key parameter, abnormality detection is performed based on a first alarm period at the beginning, that is, abnormality is detected every 30 minutes, and if the latest data value at that time is abnormal, an alarm is performed; when the alarm occurs (namely after the alarm is given for the first time), switching to abnormal detection based on a second alarm period, namely detecting the abnormality every 5 minutes, and if the latest data value at that time is abnormal, giving an alarm; and when the alarm is not needed for three times continuously in the shorter alarm period, switching to the first alarm period to perform the abnormal detection, and so on, when the alarm is more frequent, performing the abnormal detection by adopting the short period, and when the alarm is more sparse, performing the abnormal detection by adopting the long period. Under the implementation mode of the change period, after the alarm occurs, if the alarm continuously exceeds the threshold value, the parameter change condition can be monitored in real time, and if the alarm only occasionally exceeds the threshold value by a certain number, the parameter change condition can be immediately recovered to a long period, so that frequent early warning is avoided.

The fluctuation range limiting alarm strategy is to alarm when the fluctuation value of the monitoring parameter data in a specified period exceeds a fluctuation range threshold value. The policy may be implemented, for example, as: when an alarm strategy is limited by a fluctuation range, alarm strategy parameter information of an alarm period and the fluctuation range is set, after an identifier of the alarm strategy is obtained, parameters of the alarm period and the fluctuation range are obtained, timing is carried out according to the alarm period, when a time interval of the alarm period is reached, a maximum value and a minimum value in a monitoring parameter data record in the time interval range of the alarm period are obtained from a storage unit, a fluctuation value of the monitoring parameter is calculated according to the maximum value and the minimum value, the fluctuation value is compared with a threshold value limited by the fluctuation range, and when the fluctuation value exceeds the threshold value limited by the fluctuation range, alarm is carried out. Taking the liquid helium pressure of the nmr as an example, the specific implementation of the strategy may be, for example: setting a fluctuation range limiting alarm strategy for key parameters of liquid helium pressure of a certain nuclear magnetic resonance device, setting an alarm period of 30 minutes and a fluctuation range of 10%, obtaining the alarm strategy for judgment, obtaining the alarm period for timing when judging that the fluctuation range limiting alarm strategy is adopted, carrying out anomaly detection every 30 minutes according to a timing result, specifically, obtaining a maximum value Amax and a minimum value Amin of liquid helium pressure data within the 30 minutes from a storage unit, subtracting the maximum value Amax and the minimum value Amin to obtain an absolute value which is a fluctuation value, then obtaining a fluctuation range, an upper threshold and a lower threshold corresponding to the liquid helium pressure of the device, calculating the fluctuation range threshold of the liquid helium pressure of the device to be (upper threshold-lower threshold) × 10%, then comparing the fluctuation value with the fluctuation range threshold, and if the fluctuation value is larger than the fluctuation range threshold, that is, | Amax-Amin | > (upper limit threshold value-lower limit threshold value) × 10%, the occurrence time of the maximum value is taken as an early warning point, and an alarm is given.

Step S302: and acquiring the liquid helium pressure value of the equipment and the occurrence time of the liquid helium pressure value in real time.

The implementation method of this step is the same as step S102 in fig. 1, and reference is made to the foregoing description, which is not repeated herein.

Step S303: and judging whether the liquid helium is abnormal or not according to the real-time liquid helium pressure value and a preset liquid helium pressure alarm threshold value, and generating an abnormal data information record according to a judgment result.

The implementation method of this step can refer to the implementation manner of step S103, which is not described herein.

Step S304: and judging the early warning value according to the warning strategy to obtain the latest early warning point.

And acquiring an alarm strategy, judging whether the current abnormal value is an early warning value or not based on the alarm strategy, namely judging whether the current abnormal value needs to be early warned or not according to the alarm strategy, when the judgment result is that the current abnormal value is the early warning value, performing step S305, calculating the residual emergency maintenance time, and when the judgment result is that the current abnormal value is a non-early warning value, continuously performing the above abnormity analysis on the next real-time liquid helium pressure value. The specific implementation method for judging whether the current abnormal value is the early warning value according to the alarm strategy may be as follows: firstly, according to the currently detected equipment ID, an alarm strategy corresponding to the liquid helium pressure of the current equipment is obtained from an abnormal configuration information database, then, early warning point detection is carried out according to the alarm strategy, and whether an early warning value exists in the current abnormal value or not is judged. The method for acquiring the alarm strategy is realized through database operation, namely, the device ID is matched with the device ID in the abnormal configuration information, so that the threshold type and the alarm strategy are acquired, wherein the threshold type corresponds to the corresponding monitoring parameter of the device, and the embodiment of the invention is liquid helium pressure. And the processing of corresponding early warning value judgment is carried out according to the alarm strategy, specifically, the judgment is carried out according to the obtained identification content of the alarm strategy, and the detection processing of the early warning value of the corresponding strategy is executed according to the judgment result, which includes the following three conditions:

in case one, if the acquired alarm policy content is the "continuous deterioration alarm policy", the maximum value of the liquid helium pressure in the current equipment on the current day by the current date is acquired from the abnormal data information record of the liquid helium pressure in the current equipment, specifically, the maximum value of the liquid helium pressure in the current equipment on the current day by the current abnormal data information record may be acquired through traversing the abnormal data information record generated in step S303 and through a database retrieval formula. And then judging whether the maximum value is an early-warning value or not, if so, not performing early warning again, so that the calculation of the remaining emergency maintenance time is not performed, and if not, the value is an abnormal value which is continuously deteriorated, namely, the value is used as an early warning value to perform early warning, and at this moment, performing step S305 to calculate the remaining emergency maintenance time. The judgment of whether repeated alarm is performed may be to store information of each alarm, perform matching judgment when the maximum value is obtained, or perform status marking on the alarm record. In the implementation mode of determining the latest early warning point by comparing the current maximum value cut off on the same day with the early warning value, the current extreme value on the same day is detected every time, so that the strategy requirement of performing early warning only by continuous deterioration can be met, and the real-time requirement of abnormal warning can be met because the abnormal detection and the judgment on the early warning value are based on real-time data, so that the alarm noise is reduced under the condition of not compromising the real-time performance, and the defect of calculating the residual emergency maintenance time for many times is avoided.

And in the second case, if the identification content of the alarm strategy is the 'period limited alarm strategy', acquiring an upper limit threshold, a lower limit threshold and an alarm period corresponding to the liquid helium pressure from the abnormal configuration information database according to the equipment ID of the currently detected equipment. And (3) acquiring the latest liquid helium pressure data value from the liquid helium pressure information database at that time according to the alarm period, judging whether the current equipment is abnormal according to the latest data value, the upper limit threshold value and the lower limit threshold value, if so, taking the latest data value at that time as an early warning value, and performing step S305 to calculate the remaining emergency maintenance time. Specifically, after the alarm period is acquired, abnormality detection is performed based on the alarm period, that is, abnormality detection is performed once per alarm period, and if the current liquid helium pressure data value exceeds the threshold value, it is determined that the liquid helium pressure data value is abnormal. For example, if the alarm period for acquiring the liquid helium pressure of the nuclear magnetic resonance apparatus with the apparatus ID MR _000af77b8aa5 is 30, the liquid helium pressure information database is read once every 30 minutes, the latest liquid helium pressure data value is acquired therefrom, the acquired latest liquid helium pressure data value is compared with the upper limit threshold and the lower limit threshold respectively, and if the latest data value is greater than the upper limit threshold or less than the lower limit threshold, the latest data value is determined as the early warning value. The calculation mode of the time interval of each alarm period can be realized by setting a timer for timing.

Case three: and if the identification content of the alarm strategy is a fluctuation range limited alarm strategy, acquiring an upper limit threshold value, a lower limit threshold value, an alarm period and a fluctuation range corresponding to the liquid helium pressure value from the abnormal configuration information database. And timing according to the alarm period, when the time interval of the alarm period is reached, acquiring the maximum value and the minimum value of the liquid helium pressure in the alarm period time interval from the liquid helium pressure information database, and performing difference on the maximum value and the minimum value to obtain the absolute value of the difference value, wherein the absolute value is used as the fluctuation value of the liquid helium pressure in the alarm period time range. And after obtaining the fluctuation value, calculating the fluctuation range threshold value of the liquid helium pressure according to the obtained upper limit threshold value, lower limit threshold value and fluctuation range of the liquid helium pressure by a formula (upper limit threshold value-lower limit threshold value) and the fluctuation range. For example, the alarm period, the upper threshold, the lower threshold and the fluctuation range of the liquid helium pressure of the nuclear magnetic resonance equipment with the equipment ID of MR _000af77b8aa5 are respectively: the alarm period is 10, the upper threshold is 4.95, the lower threshold is 3.85, and the fluctuation range is 10%. Therefore, early warning detection is carried out on the liquid helium according to the warning period of the liquid helium pressure, and specifically, a fluctuation value is calculated every 10 minutes: and acquiring the maximum value and the minimum value in the liquid helium pressure information database within 10 minutes, taking the maximum value-the minimum value and the absolute value to obtain the fluctuation value of the liquid helium pressure within 10 minutes, and acquiring the fluctuation range threshold value of (upper limit threshold value-lower limit threshold value) × 10%, namely (4.95-3.85) × 10%. And then, comparing the calculated fluctuation value with a fluctuation range threshold value, if the fluctuation value exceeds, namely is larger than, the fluctuation range threshold value, judging that the current point is an early warning point, and at the moment, performing step S305 to calculate the remaining emergency maintenance time.

Step S305: and generating the residual emergency maintenance time according to the abnormal data information record and the preset leakage threshold value.

The mode of calculating the remaining emergency maintenance time according to the abnormal data information record and the preset leakage threshold value is also realized through curve fitting. The curve fitting may be a linear fitting or a non-linear fitting, wherein an implementation of the least square normal fitting is preferred in the embodiment of the present invention, and a specific implementation thereof can be shown in fig. 2 and described in the foregoing, and is not described herein again. The difference between this step of the embodiment of the present invention and the corresponding step of fig. 1 is that this step is performed when the early warning point is detected according to the alarm strategy, and in terms of the selection of the input value, the embodiment of the present invention may use all data in the abnormal data information record when the latest early warning point is detected as the input value of curve fitting, so as to obtain the slope v of the fitting parameter, i.e., the change rate of the liquid helium pressure; or taking the abnormal data record near the detected latest early warning point (such as within 10 minutes or 20 minutes before the early warning point) as an input value of curve fitting, thereby obtaining the slope v of the fitting parameter. For the continuous deterioration alarm strategy, when the current maximum value in the detected abnormal record on the same day is a non-early-warning value, the maximum value point is the latest early-warning point; for the period limit alarm strategy, when the detected current latest data value is an abnormal value, the latest data value point is the latest early warning point; and limiting an alarm strategy for the fluctuation range, and when the fluctuation value in the current alarm period is detected to be larger than the threshold value of the fluctuation range, determining the maximum value point in the alarm period as the latest early warning point. After determining the input values for the curve fit, other calculations may be performed with reference to the method shown in FIG. 2 and described above.

It should be noted that, in the period-limited alarm strategy, the latest data value may be an abnormal value, or may be a case where the latest data value is smaller than the lower threshold, in this case, the occurrence time of the abnormal value may be taken to select a nearby abnormal point, or a next highest abnormal value point (i.e., a value point where the abnormal value is larger than the upper threshold) closest to the abnormal value may be taken as a latest early-warning point, so as to obtain the abnormal value near the highest abnormal value point to estimate the remaining emergency repair time. In a preferred embodiment, the period-limited alarm strategy may only make an estimate of the remaining emergency repair time for situations greater than an upper threshold.

Step S306: and generating alarm information according to the latest early warning point and the generated residual emergency maintenance time and outputting the alarm information.

After the remaining emergency maintenance time is obtained, generating alarm information according to the information of the latest early warning point and the remaining emergency maintenance time detected by the alarm strategy, and outputting the alarm information, wherein the alarm information can be set to include an equipment ID, a liquid helium pressure value and occurrence time of the early warning point, a leakage threshold, time for reaching the leakage threshold from the latest early warning point according to the current trend, and the remaining emergency maintenance time, and can also include other contents such as an equipment name or an address.

By the method, the residual maintenance time can be estimated based on the real-time liquid helium pressure value of the nuclear magnetic resonance equipment and the alarm strategy, and the alarm is performed according to the alarm strategy set by a user after the residual time is estimated. Therefore, accurate estimation can be carried out according to the user requirements and the actual conditions, and the estimation result is timely reported to the user, so that the residual time is informed to the user in advance before leakage is possible, the user can conveniently take measures in time, and loss is avoided. Moreover, the early warning does not need manual observation of the pressure condition of the liquid helium of the equipment, so that the labor cost is reduced, and the efficiency is improved. In other embodiments, the nuclear magnetic resonance apparatus may be extended to other apparatuses, and the liquid helium pressure may be extended to other monitoring parameters, and the specific implementation manner of the alarm threshold, the greater hazard threshold, and the alarm strategy may be adaptively changed based on the basic concept according to the characteristics of the apparatuses and the monitoring parameters.

In the embodiment of the present invention, generating the alarm information output refers to generating information output including the device identifier, the liquid helium pressure value of the early warning point, the occurrence time of the liquid helium pressure value, the leakage threshold value, and the remaining emergency maintenance time, where the information output may be output to a user page, or may be sent to a user by a mobile phone short message or a mail (a mobile phone number and a mail address may be entered when an account is assigned), or may be output to a user terminal, and the user terminal may be, for example, a WeChat or a client APP corresponding to the system of the embodiment of the present invention. Fig. 4 schematically shows a specific implementation manner of outputting alarm information to a user terminal, taking a user terminal as an example of WeChat, as shown in fig. 4, the method includes:

step S401: account information is assigned to the user and the device is associated with the account information.

The account information is distributed for each user, the equipment responsible for the user is appointed for the user distributed with the account, the account information comprising a user name and a password can be added through a user page, the equipment ID related to each account is distributed through the page, and then the user information database comprising the user name, the password and the equipment ID is generated according to the input of the page.

Step S402: the account information of the user is associated with the WeChat of the user.

The user notes the public number through the WeChat, enters the page of the system through the public number and carries out the binding operation, wherein, carrying out the binding operation includes: firstly, logging in a system from a system page of a public number through the user name and the password distributed in the step S401, then sending a binding request to the system, obtaining the binding request by the system, analyzing the user name, the password and the micro-signal in the binding request, firstly, carrying out identity verification according to the user name and the password, and after the verification is passed, associating the account information in the request with the micro-signal account number of the user, namely, generating a user information database comprising the user name, the password, the equipment ID and the micro-signal.

Step S403: and outputting the generated alarm information to a WeChat terminal of a corresponding user.

And then, after the alarm information is generated, the system acquires the micro signal bound by the user from the user information database according to the equipment ID needing alarming, and sends the generated alarm information to a micro signal terminal corresponding to the micro signal.

Through the steps, the user can bind the WeChat with the method and the system, so that the alarm prompt is sent to the communication tool commonly used by the user according to the habit of the user, the user is prompted in time, the user experience is improved, and the method and the system are efficient and quick.

Fig. 5 schematically shows a frame structure of a remaining emergency repair time estimation system for equipment according to an embodiment of the present invention. As shown in fig. 5, the estimation system 1 includes a storage unit 10, an information configuration module 11, an abnormality data collection module 12, and a remaining emergency maintenance time estimation module 13. The information configuration module 11 is configured to receive a device monitoring parameter, an alarm threshold and a larger hazard threshold set by a user and store the device monitoring parameter, the alarm threshold and the larger hazard threshold in the storage unit 10, the abnormal data collection module 12 is configured to obtain real-time monitoring parameter data, judge whether an abnormality occurs according to a real-time monitoring parameter data value and a preset alarm threshold, collect abnormal data information according to a judgment result, generate an abnormal data information record and store the abnormal data information record in the storage unit 10, and the remaining emergency maintenance time estimation module 13 is configured to generate remaining emergency maintenance time according to the abnormal data information record and the preset larger hazard threshold. In a specific implementation, the storage unit 10 may be a database, or may be other storage manners such as a file storage system or a custom storage structure, and the database storage manner is preferred in the embodiment of the present invention. A user or an administrator sets an alarm threshold and a larger hazard threshold through a page, the information configuration module 11 receives user input and stores the user input, and the stored information comprises an equipment ID, an upper limit threshold, a lower limit threshold and a larger hazard threshold. The abnormal data collecting module 12 obtains the monitoring parameter data information in real time to compare with the alarm threshold, and when the monitoring parameter data information exceeds the alarm threshold and develops toward a larger damage threshold (for example, the larger damage threshold occurs along with the increase of the monitoring parameter data, and the abnormal condition larger than the upper threshold develops toward the larger damage threshold at this time), and conversely, if the larger damage threshold occurs along with the decrease of the monitoring parameter data, and the abnormal condition smaller than the lower threshold develops toward the larger damage threshold at this time), obtains the monitoring parameter data information to generate and store an abnormal data information record, wherein the abnormal data information record includes the equipment ID, the abnormal key parameter value and the occurrence time. The abnormal data collection module 12 may obtain the real-time key parameter data from the storage unit 10 or read the data from the device in real time, and the real-time data in the storage unit 10 may be recorded in real time, or recorded at regular time, or collected from the device in real time. When an abnormality developing toward a larger hazard threshold occurs, the remaining emergency maintenance time estimation module 13 obtains the abnormal data information record, and calculates the remaining emergency maintenance time output in a curve fitting manner, where the specific calculation manner may refer to the foregoing description, and this embodiment is not described herein again. By the system provided by the embodiment of the invention, the real-time key parameter data of the equipment can be utilized to predict the deterioration condition of the monitoring parameter of the equipment, such as liquid helium, and predict the residual emergency maintenance time, namely the time when the current monitoring parameter value reaches a larger damage threshold value according to the current deterioration speed distance, the calculation result is accurate, the processing speed is high, and efficient and automatic prediction can be realized.

Fig. 6 schematically shows a remaining emergency repair time early warning system for equipment according to an embodiment of the present invention, and as shown in fig. 6, the early warning system 2 includes a storage unit 20, a parameter setting module 21, a remaining emergency repair time estimation module 22, and an alarm module 23. The parameter setting module 21 is configured to receive user input, so as to set corresponding parameter information according to a requirement, where the parameter information set by the user may include a monitoring parameter, an alarm threshold (including an upper threshold and a lower threshold), and a greater hazard threshold, and in other embodiments, the set parameter information may further include an alarm policy. The parameter setting module 21 receives parameter information set by a user for each device, and stores the device ID and corresponding parameter information, such as the device ID, the upper threshold, the lower threshold, and the leakage threshold, in the storage unit 20. Then, the remaining emergency repair time estimation module 22 obtains the real-time monitoring parameter value to perform the processing of estimating the remaining emergency repair time, and outputs the remaining emergency repair time to the alarm module 23 after the processing. After receiving the remaining emergency maintenance time, the alarm module 23 generates an alarm information output including the current device ID, the current key parameter value, the larger hazard threshold value, and the remaining emergency maintenance time, where the output may be output to a user page, a mobile phone short message of the user, a mail, or a communication application software on an intelligent terminal of the user. The remaining emergency repair time estimation module 22 obtains the real-time key parameter value to perform a specific processing method of estimating the remaining emergency repair time, which can be referred to the above description.

Fig. 7 schematically shows a remaining emergency repair time early warning system for equipment according to another embodiment of the present invention, which is illustrated by taking as an example that liquid helium pressure data of a nuclear magnetic resonance device is collected in real time (that is, the equipment is a nuclear magnetic resonance device, and a monitoring parameter is liquid helium pressure), and alarm information is received by a user terminal, as shown in fig. 7, the system 2 includes an information collection module 25, a user terminal 26, and a system platform 24, where the information collection module 25 is connected to each nuclear magnetic resonance device 1, and is configured to obtain real-time liquid helium pressure data output by the device 1 and occurrence time thereof in real time and upload the data to the system platform 24, and the system platform 24 and the user terminal 26 are in mutual communication, and are configured to output the alarm information to the user terminal 26. The system platform 24 in the embodiment of the present invention includes a storage unit 240, a parameter setting module 241, a remaining emergency maintenance time estimation module 242, a user management module 244, a terminal binding module 245, and an alarm module 243, where the parameter setting module 241 is configured to receive parameter information input by a user and store the parameter information in the storage unit 240, and includes a liquid helium pressure alarm threshold and a larger hazard threshold (i.e., a leakage threshold), the remaining emergency maintenance time estimation module 242 is configured to perform analysis processing according to liquid helium pressure information data uploaded by the information acquisition module 25 in real time, and generate a remaining emergency maintenance time output, and the alarm module 243 is configured to generate alarm information including an equipment ID, a current liquid helium pressure value, a leakage threshold, and a remaining emergency maintenance time according to the equipment information and the remaining emergency maintenance time and output the alarm information to the user terminal 26. The remaining emergency maintenance time estimation module 242 includes an abnormal data collection unit and a time calculation unit (not shown), where the abnormal data collection unit is configured to obtain the liquid helium pressure information data uploaded by the information collection module 25 in real time, determine whether an abnormality occurs, and generate an abnormal data information record according to a determination result; and the time calculation unit is used for generating the residual emergency maintenance time according to the generated abnormal data information record and the preset leakage threshold value. The processing method of the abnormal data collecting unit for specifically performing the abnormal judgment and generating the abnormal data information record can be referred to the above description, and similarly, the method of the time calculating unit for generating the remaining emergency maintenance time can be referred to the above description.

In order to implement the alarm module 243 to directly output the alarm information to the user terminal 26 in the embodiment of the present invention, the user terminal 26 needs to be bound with the system platform 24 first when the alarm module is used specifically. Specifically, the user terminal 26 needs to be associated with each user equipment through the user management module 244 and the terminal binding module 245, specifically: the user management module 244 is configured to allocate account information for the user and associate the allocated account information with corresponding device information, and the terminal binding module 245 is configured to receive a binding request sent by the user terminal 3 and associate the user terminal 3 with the account information allocated by the user management module 244 according to the binding request. The account information allocated by the user management module 244 to the user includes a user name and a password, the account information is generated as a user information database and stored in the storage unit 240, and associating the allocated account information with the corresponding device information means that a device ID associated with the user name is set in a record corresponding to the user name to designate each user as its allocated device, so that the alarm information of the device can be sent to the designated user, and the designated user performs corresponding processing, such as maintenance and the like, on the device. The terminal binding module 245 obtains account information and terminal information of the user from the binding request sent by the user from the user terminal 3 of the user (the user needs to input a user name and a password when the user performs binding through the user terminal 3), and performs identity authentication according to the account information to associate and bind each user with the corresponding user terminal 3 when the authentication passes, where the association binding may be to store the obtained user terminal information in a user information database record of the corresponding user, for example, to generate a user information database record including a user name-password-associated device ID-associated user terminal ID database structure. By the system provided by the embodiment of the invention, account information can be distributed to the user according to requirements and the administered equipment can be appointed for the corresponding user, and then the user can bind the commonly used user terminal to the system platform provided by the embodiment of the invention according to the convenience of the user, so that the alarm information can be output to the bound user terminal, the user can check the alarm information at any time conveniently, and the equipment can be monitored and maintained in time. Thus, when an abnormality occurs, the alarm module 243 can find the corresponding user and the user terminal thereof according to the abnormal device ID, so as to output the generated alarm information to the user page or the user terminal of the user, thereby outputting the alarm prompt of the corresponding device to the designated user. The system of the embodiment of the invention realizes the communication between each device and the user terminal at the same time, and provides a more convenient implementation scheme for monitoring parameter values and early warning of the remaining emergency maintenance time. The specific implementation manner of the parameter setting module 241 in the embodiment of the present invention may refer to the foregoing description, and the information collecting module 25 may be implemented by an existing data collecting box, where the collected data may be sensor data or an equipment log file.

Fig. 8 schematically shows a frame structure of a remaining emergency repair time early warning system for equipment in still another embodiment of the present invention. As shown in fig. 8, the system according to the embodiment of the present invention is similar to the system shown in fig. 7, except that the parameter information set by the parameter setting module in the embodiment of the present invention further includes an alarm policy, the system platform according to the embodiment of the present invention further includes an early warning point obtaining module 247, the information collecting module 25 according to the embodiment of the present invention is configured to obtain a stored device log from a nuclear magnetic resonance device connected to the information collecting module, and correspondingly, the system according to the embodiment of the present invention further provides a device log analyzing module 246 in the system platform 24, so as to analyze the uploaded device log. The device log in this embodiment may be in the form of a file stored in a storage system of the device itself, or in other forms, as long as the device log is continuously updated and in an extractable state as the device is used, and the device log can be accessed and utilized by an external device, which is not limited in this embodiment of the present invention. The data collection module 25 in the embodiment of the present invention may be a data collection box, which is directly disposed on a device requiring device log collection, reads the device log on the device through a data collection function of the data collection box, and sends the device log to the system platform 24 through a network. The system platform 24 may be implemented by software programming and deployed on a server (which may be deployed on a cloud server or a server local to a hospital). After the system platform 24 receives the device log through the network, the device log is first stored in the storage unit 240, and then the device log parsing module 246 traverses the device log in the storage unit 240 to parse the device log. The method for analyzing may be to extract required key information, such as a liquid helium pressure value and occurrence time (i.e., a timestamp) thereof, through a feature matching algorithm (e.g., a screening algorithm based on a regular expression), where the key information may be customized according to needs and types of devices, but at least needs to include the occurrence time and a monitoring parameter value. And after the key information is analyzed, extracting the key information and storing the extracted key information as a data record of the database according to a minimum index unit of 'equipment ID or name-occurrence time-monitoring parameter value'. The remaining emergency maintenance time estimation module 242 obtains a monitoring parameter value from a monitoring parameter information record of the database, such as a liquid helium pressure information data record, in real time through the abnormal data collection unit, compares the monitoring parameter value with a preset alarm threshold value, determines whether an abnormality occurs according to a comparison result, and generates and stores an abnormal data information record when the abnormality occurs, wherein the abnormality stored in the abnormal data information record needs to be an abnormality which develops toward a direction of a larger damage threshold value. The early warning point acquisition module acquires early warning strategy information for judgment (for example, when an abnormality occurs or new real-time monitoring parameter data is acquired), so as to perform early warning detection according to an alarm strategy, determine a latest early warning point according to an alarm threshold, a larger hazard threshold, early warning strategy information and monitoring parameter data records under different alarm strategies, and notify the time calculation unit to calculate the remaining emergency maintenance time when the latest early warning point is determined. When the latest early warning point is determined, the time calculation unit generates the remaining emergency maintenance time according to the abnormal data information record and the curve fitting method and outputs the remaining emergency maintenance time to the alarm module 243. The alarm module 243 outputs alarm information including the device ID, the latest early warning point monitoring parameter value and the occurrence time thereof, the greater hazard threshold value, and the remaining emergency maintenance time, generated based on the latest early warning point information and the remaining emergency maintenance time, to the user terminal 26. The specific implementation processes of the early warning point acquisition module and the time calculation unit may refer to the description of the method part. Therefore, monitoring analysis of monitoring parameter data of equipment and estimation and early warning of the remaining emergency maintenance time can be carried out through the system of the embodiment of the invention, the data acquisition module in the system of the embodiment of the invention is directly connected with corresponding equipment, equipment logs are acquired, log analysis and abnormity analysis are carried out by a system platform, the remaining emergency maintenance time is calculated under the condition that an alarm strategy is met, alarm information is generated and output to a corresponding user terminal, the maintenance time is not judged manually, and the labor cost is reduced. In addition, the time estimation of the embodiment of the invention is to analyze and calculate according to the threshold set aiming at different equipment conditions and the real-time analysis result of the equipment log, so that the estimation result is more accurate, and the loss can be greatly avoided by early warning.

Those skilled in the art can understand that the output terminal in the embodiment of the present invention may be only one of WeChat, SMS, email, or client APP, or may be any combination of two or more.

It should be noted that, the embodiment of the present invention does not limit the manner of acquiring the monitoring parameter information data of the device in real time, and no matter what manner, the present invention only needs to acquire the real-time monitoring parameter data of the device. By the method and the system, all medical equipment in each hospital can be monitored and early warned, and all medical equipment in each hospital can be monitored and early warned through a cloud system, so that the time for reaching a larger harm threshold value can be predicted more quickly and accurately according to the deterioration condition and the larger harm threshold value and the alarm can be given timely and effectively, and the efficiency is improved effectively as long as the monitoring parameter information data of each equipment can be acquired and the threshold value and the alarm strategy are set according to the condition of each equipment.

In addition, the method and system of the embodiment of the present invention can also be implemented by an electronic device, which only needs to include a storage medium and a control unit, and stores the application program that implements the method or system of the embodiment of the present invention in the storage medium, and the control unit executes the application program to achieve the purpose of the present invention. Such an electronic device may be, for example, an intelligent terminal device (e.g., a smart phone or a smart watch) capable of carrying and executing an application program, or a tablet computer.

What has been described above are merely some embodiments of the present invention. It will be apparent to those skilled in the art that various changes and modifications can be made without departing from the inventive concept thereof, and these changes and modifications can be made without departing from the spirit and scope of the invention.