1. A leakage current measurement error compensation method based on cloud edge collaborative computing is characterized in that the method is realized on a communication network formed by interconnection of a leakage current edge monitoring terminal and a power utilization management cloud platform, and comprises the following steps:

the leakage current edge monitoring terminal monitors load voltage, load current and leakage current data and sends the data to the power utilization management cloud platform;

the power utilization management cloud platform iteratively trains a false leakage compensation model of the false leakage compensation model by using the received data, updates parameters of the false leakage compensation model and feeds the parameters back to the leakage current edge monitoring terminal;

and the leakage current edge monitoring terminal uses a false leakage compensation model which is the same as the power utilization management cloud platform, and processes the leakage current data according to the false leakage compensation model parameters received from the power utilization management cloud platform so as to eliminate the influence of false leakage phenomenon in the leakage current data in real time.

2. The method according to claim 1, wherein the load voltage, the load current and the leakage current data are collected from a predetermined sensor at a specific line node.

3. The leakage measurement error compensation method based on cloud-edge cooperative computing as claimed in claim 1, wherein the pseudo leakage compensation model is constructed based on a multi-model regression algorithm of robust fuzzy clustering, and the center point and covariance of the fuzzy member are gradually optimized in the training process until the probability member is converged.

4. The leakage measurement error compensation method based on cloud-edge cooperative computing as claimed in claim 3, wherein the fuzzy member optimization formula isWherein the content of the first and second substances,

c is the clustering number; m is>1, and is an obfuscated weight parameter; x is the number of_jIs a sample; c. C_iIs the center of cluster i; e.g. of the type_ikCovariance matrix for cluster iThe kth feature vector of (1); v. of_ikIs the weight coefficient of the feature root in the k-th feature vector.

5. The leakage measurement error compensation method based on cloud-edge cooperative computing as claimed in claim 4, wherein the update formula of the probability member is

6. A leakage current measurement error compensation system based on cloud edge collaborative computing is characterized by comprising a leakage current edge monitoring terminal and an electricity management cloud platform which are interconnected; wherein the content of the first and second substances,

the leakage current edge monitoring terminal is used for monitoring leakage current data and sending the leakage current data to the power utilization management cloud platform; processing the leakage current data by using a false leakage compensation model which is the same as the power utilization management cloud platform according to the false leakage compensation model parameters received from the power utilization management cloud platform so as to eliminate the influence of false leakage phenomenon in the leakage current data;

and the power utilization management cloud platform is used for iteratively training a pseudo leakage compensation model by using the received leakage data, updating parameters of the pseudo leakage compensation model and feeding back the parameters to the leakage current edge monitoring terminal.

7. The system according to claim 6, wherein the load voltage, the load current and the leakage current data are collected from a predetermined sensor at a specific line node.

8. The leakage measurement error compensation system based on cloud-edge cooperative computing as claimed in claim 6, wherein the pseudo leakage compensation model is constructed based on a multi-model regression algorithm of robust fuzzy clustering, and the center point and covariance of the fuzzy member are gradually optimized in the training process until the probability member is converged.

Background

With the rapid development of national economy, the generated energy and the power consumption of China leap the world first, and the frequent occurrence of electrical safety accidents is brought along with the leap of the generated energy and the power consumption of China. The most of the positions where electrical fire occurs are in the range from a power distribution end to an electricity utilization terminal in an electric power system (namely, from a user-side low-voltage transformer, through a low-voltage distribution electrical appliance and then to a user-side electrical appliance), because distribution lines and the electrical appliance are usually installed and maintained by the user, so that nonstandard phenomena often exist in the links of design, installation, use and the like, and professional maintenance and overhaul can not be carried out even after damage occurs in the long-term operation process, especially the most important reason is that the line state lacks effective monitoring. However, in the case of electrical safety accidents, the proportion of personal safety accidents and fire accidents caused by leakage currents in the circuit is large. Therefore, it is very important to monitor the leakage current in the line in real time.

At present, a zero sequence mutual current sensor is adopted to measure leakage current in a line. However, in practical application, a zero sequence transformer generates a measurement error due to magnetic leakage, uneven winding and the like, so that the measured leakage current is significantly larger than the real leakage current in a line, and is referred to as a "false leakage" phenomenon in the industry. The interference of 'false leakage' must be removed to accurately obtain the real leakage current closely related to the electrical fire.

Theoretically, the magnitude of the leakage current is not affected by the operating current variations. However, in the research, it is found that the "false leakage" increases with the increase of the working current, which is caused by the precision of the zero sequence transformer itself and the cable installation manner. For example, the characteristics of different zero sequence transformers are difficult to be completely consistent, so that the precision is inaccurate; as another example, the consistency of the cable installation cannot be guaranteed, resulting in a non-constant distribution pattern of true and "false leakages". Therefore, it is necessary to separately model the false leakage situation of each zero sequence transformer by analyzing the monitoring data of each zero sequence transformer. Due to insufficient computing power of the edge monitoring terminal, data needs to be uploaded to a cloud platform for modeling calculation, and then compensation parameters are fed back to the edge terminal for false leakage compensation.

Disclosure of Invention

The technical problem to be solved by the embodiments of the present invention is to provide a leakage measurement error compensation method and system based on cloud-edge cooperative computing, which can eliminate a false leakage phenomenon generated by a zero-sequence current transformer to obtain accurate leakage current data.

In order to solve the above technical problem, an embodiment of the present invention provides a leakage current measurement error compensation method based on cloud-edge collaborative computing, which is implemented on a communication network formed by interconnection of a leakage current edge monitoring terminal and a power management cloud platform, and includes the following steps:

the leakage current edge monitoring terminal monitors load voltage, load current and leakage current data and sends the data to the power utilization management cloud platform;

the power utilization management cloud platform iteratively trains a pseudo leakage compensation model by using the received data, updates parameters of the pseudo leakage compensation model and feeds the parameters back to the leakage current edge monitoring terminal;

and the leakage current edge monitoring terminal uses a false leakage compensation model which is the same as the power consumption management cloud platform, and processes the leakage current data according to the false leakage compensation model parameters received from the power consumption management cloud platform so as to eliminate the false leakage compensation model influenced by the false leakage phenomenon in the leakage current data.

The load voltage, the load current and the leakage current data are acquired from a preset sensor on a specified line node.

The false leakage compensation model is constructed based on a multi-model regression algorithm of robust fuzzy clustering, and the central point and covariance of the fuzzy member are gradually optimized in the training process until the probability member is converged.

Wherein the optimization formula of the fuzzy member isWherein the content of the first and second substances,

c is the clustering number; m is>1, and is an obfuscated weight parameter; x is the number of_jIs a sample; c. C_iIs the center of cluster i; e.g. of the type_ikCovariance matrix for cluster iThe kth feature vector of (1); v. of_ikIs the weight coefficient of the feature root in the k-th feature vector.

Wherein the update formula of the probability member is

The embodiment of the invention also provides a leakage current measurement error compensation system based on cloud-edge cooperative computing, which comprises a leakage current edge monitoring terminal and an electricity management cloud platform which are interconnected; wherein the content of the first and second substances,

the leakage current edge monitoring terminal is used for monitoring load voltage, load current and leakage current data and sending the data to the power utilization management cloud platform; processing the leakage current data by using a false leakage compensation model which is the same as the power utilization management cloud platform according to the false leakage compensation model parameters received from the power utilization management cloud platform so as to eliminate the influence of false leakage phenomenon in the leakage current data;

the power utilization management cloud platform is used for iteratively training a false leakage compensation model by using the received data, updating parameters of the false leakage model and feeding back the parameters to the false leakage compensation model of the leakage current edge monitoring terminal.

The load voltage, the load current and the leakage current data are acquired from a preset sensor on a specified line node.

The false leakage compensation model is constructed based on a multi-model regression algorithm of robust fuzzy clustering, and the central point and covariance of the fuzzy member are gradually optimized in the training process until the probability member is converged.

The embodiment of the invention has the following beneficial effects:

based on the original storage function of the power utilization management cloud platform, the intelligent closed-loop feedback of parameter self-adaption and self-adjustment is realized by focusing on solving the problems of construction and training of a complex model, model parameter updating is provided for edge calculation and is fed back to the leakage current edge monitoring terminal, so that the leakage current edge monitoring terminal processes data acquired by the zero-sequence current transformer according to the false leakage model parameters to complete edge calculation compensation, the influence of false leakage phenomena on real leakage current is eliminated, and the purpose of eliminating the false leakage phenomena generated by the zero-sequence current transformer and obtaining accurate leakage current data is achieved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is within the scope of the present invention for those skilled in the art to obtain other drawings based on the drawings without inventive exercise.

Fig. 1 is a flowchart of a leakage measurement error compensation method based on cloud-edge collaborative computation according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a leakage measurement error compensation system based on cloud-edge collaborative calculation according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings.

As shown in fig. 1, in the embodiment of the present invention, a leakage current measurement error compensation method based on cloud-edge cooperative computing is implemented on a communication network formed by interconnection of a leakage current edge monitoring terminal and a power management cloud platform, and includes the following steps:

step S1, the leakage current edge monitoring terminal monitors load voltage, load current and leakage current data and sends the data to the power utilization management cloud platform;

step S2, the power consumption management cloud platform uses the received data to iteratively train a false leakage compensation model, updates false leakage model parameters and feeds back the parameters to the leakage edge monitoring terminal;

and step S3, the leakage current edge monitoring terminal uses a false leakage compensation model which is the same as the power consumption management cloud platform, and the leakage current data are processed according to the false leakage compensation model parameters received from the power consumption management cloud platform, so that the false leakage compensation model influenced by the false leakage phenomenon in the leakage current data is eliminated.

In step S1, the leakage current edge monitoring terminal monitors load voltage, load current, and leakage current data collected from a preset sensor at a specified line node, and sends the data to the power consumption management cloud platform through the communication network. It should be noted that the communication system of the communication network includes, but is not limited to, 5G, 4G, NB-IoT, LoRaWAN, WiFi, ethernet, and third-party customized communication system.

It is understood that the leakage current edge monitoring terminal and various sensors (including a zero sequence current transformer for monitoring leakage current) may be integrated or may be separate components.

In step S2, since the interference removal of the leakage current data in the leakage current edge monitoring terminal needs to separate the real leakage current from the "false leakage" current, and the specific distribution model of the two is unknown, it needs to be self-determined in the operation data accumulation, so that the problem is an unsupervised adaptive clustering problem. Also, since both the true leakage current and the "false leakage" current are different distribution models, the above problem can be considered as a Multiple Regression (MIR) problem.

The MIR problem (RFC-MIR) is realized by adopting a Robust Fuzzy Clustering for Multiple Instance Regression in a power consumption management cloud platform, namely a false leakage compensation model is constructed based on a multi-model Regression algorithm of Robust Fuzzy Clustering, and the training of the model is completed by continuously optimizing the central point and covariance of a Fuzzy member in the training process until the probability member is converged.

Wherein, the optimization formula of the fuzzy member is shown as the following formula (1):

wherein the content of the first and second substances,c is the clustering number; m is>1, and is an obfuscated weight parameter; x is the number of_jIs a sample; c. C_iIs the center of cluster i; e.g. of the type_ikCovariance matrix for cluster iThe kth feature vector of (1); v. of_ikIs the weight coefficient of the feature root in the k-th feature vector.

Wherein, the updating formula of the probability member is shown as the following formula (2);

firstly, constructing a false leakage compensation model in a power utilization management cloud platform; secondly, a line sensing database is established in the power utilization management cloud platform, data uploaded by each topological node line monitoring unit every time are recorded, a pseudo leakage compensation model is trained according to the collected historical data, and training of the model is completed by continuously optimizing the central point and covariance of a fuzzy member in the training process until a probability member is converged, so that a trained pseudo leakage compensation model is obtained; and finally, processing leakage current data uploaded by the leakage current edge monitoring terminal by adopting a trained false leakage compensation model to obtain false leakage model parameters and feed the false leakage model parameters back to the leakage current edge monitoring terminal, so as to realize intelligent closed-loop feedback of parameter self-adaptation and self-adjustment.

In step S3, the leakage current edge monitoring terminal uses the same pseudo leakage compensation model as the power management cloud platform and processes the leakage current data according to the pseudo leakage compensation model parameters received from the power management cloud platform to eliminate the influence of the pseudo leakage phenomenon in the leakage current data. It can be understood that the leakage current data is different from historical data adopted in the training of the electricity management cloud platform, namely newly collected leakage current data.

As shown in fig. 2, in the embodiment of the present invention, the leakage current measurement error compensation system based on cloud-edge cooperative computing includes a leakage current edge monitoring terminal 1 and an electricity management cloud platform 2 that are interconnected; wherein the content of the first and second substances,

the leakage current edge monitoring terminal 1 is used for monitoring load voltage, load current and leakage current data and sending the data to the electricity management cloud platform 2; the false leakage compensation model which is the same as the power consumption management cloud platform 2 is used, and the false leakage compensation model parameters fed back by the power consumption management cloud platform 2 are received to process the leakage current data so as to eliminate the influence of the false leakage phenomenon in the leakage current data;

and the power utilization management cloud platform 2 iteratively trains the false leakage compensation model by using the received data to obtain false leakage model parameters and feeds the false leakage model parameters back to the leakage current edge monitoring terminal 1.

The load voltage, the load current and the leakage current data are acquired from a preset sensor on a specified line node.

The false leakage compensation model is constructed based on a multi-model regression algorithm of robust fuzzy clustering, and the central point and covariance of the fuzzy member are gradually optimized in the training process until the probability member is converged.

The embodiment of the invention has the following beneficial effects:

based on the original storage function of the power utilization management cloud platform, the intelligent closed-loop feedback of parameter self-adaption and self-adjustment is realized by focusing on solving the problems of construction and training of a complex model, model parameter updating is provided for edge calculation and is fed back to the leakage current edge monitoring terminal, so that the leakage current edge monitoring terminal processes data acquired by the zero-sequence current transformer according to the false leakage model parameters to complete edge calculation compensation, the influence of false leakage phenomena on real leakage current is eliminated, and the purpose of eliminating the false leakage phenomena generated by the zero-sequence current transformer and obtaining accurate leakage current data is achieved.

It should be noted that, in the above system embodiment, each included unit is only divided according to functional logic, but is not limited to the above division as long as the corresponding function can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.

It will be understood by those skilled in the art that all or part of the steps in the method for implementing the above embodiments may be implemented by relevant hardware instructed by a program, and the program may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc.

While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.