1. A safety production area personnel management method based on big data analysis is characterized by comprising the following steps:

acquiring a motion state sequence of equipment when a person arrives at a safe production area and a danger degree change sequence of the equipment when the person arrives beside the equipment; analyzing the association relation between the motion state sequence and the danger degree change sequence to obtain the association characteristics of the personnel and the equipment;

obtaining the motion state distribution characteristics of the personnel and the equipment according to the motion states of the personnel at the positions in the neighborhood of the equipment:

acquiring the association characteristics of a person and all equipment to obtain an association vector of the person; forming a first set by the association vectors of all the persons, and obtaining unit vectors in the principal component direction of the first set to obtain a first space;

transforming the vectors in the first set to a first space to obtain a second set; setting target personnel, and obtaining the replaceability of the target personnel in the principal component direction according to the direction similarity between the vector corresponding to the target personnel in the second set and the unit vector in the principal component direction and the projection length of each vector of the second set in the unit vector in the principal component direction; obtaining the replaceability of the target personnel according to the replaceability of the target personnel in the principal component direction and the motion state distribution characteristics of the target personnel and the equipment; and carrying out safety production area personnel management according to the replaceability of the target personnel.

2. The method of claim 1, further comprising:

obtaining a unit vector of a space where the first set is located to obtain a second space, and transforming the unit vector in the principal component direction of the first space to the second space, wherein a transformation coefficient corresponding to the unit vector of the second space is a weight coefficient in the principal component direction; and weighting and superposing the motion state distribution characteristics of the target person and all the devices by using the weight coefficients in the principal component directions, and obtaining the substitutability of the target person according to the substitutability of the target person in all the principal component directions and the superposition result.

3. The method of claim 1, further comprising:

and acquiring the motion state distribution of the personnel at each position of the equipment neighborhood when the personnel arrive at the equipment every time, and weighting and superposing the motion state distribution to obtain the motion state distribution characteristics of the personnel and the equipment.

4. The method according to claim 3, wherein the step of weighted superposition of motion state distributions comprises:

and determining a weight coefficient according to the variation of the danger degree of the equipment when the personnel arrive at the equipment every time, and performing weighted superposition on the motion state distribution according to the weight coefficient to obtain the motion state distribution characteristics of the personnel and the equipment.

5. The method of claim 1, further comprising:

and analyzing the association relation between the motion state sequence and the danger degree change sequence by utilizing a time sequence convolution network to obtain an association grade as the association characteristic of the personnel and the equipment.

6. The method of claim 1, further comprising:

acquiring the association relationship between the motion state sequence of the equipment when the personnel arrives at the safe production area and the danger degree change sequence of the equipment when the personnel arrives beside the equipment, and acquiring the association characteristics of the personnel and the equipment according to the average value of all the association relationships.

7. A safety production area personnel management system based on big data analysis is characterized by comprising:

the data acquisition module is used for acquiring a motion state sequence of equipment when personnel arrive at a safe production area and a danger degree change sequence of the equipment when the personnel arrive beside the equipment;

the system comprises a relation characteristic acquisition module, a risk degree change sequence acquisition module and a risk degree change sequence acquisition module, wherein the relation characteristic acquisition module is used for analyzing the relation between the motion state sequence and the risk degree change sequence to obtain the relation characteristics of personnel and equipment; obtaining the motion state distribution characteristics of the personnel and the equipment according to the motion state of the personnel at each position in the neighborhood of the equipment;

the data analysis module is used for acquiring the association characteristics of the personnel and all the equipment to obtain the association vector of the personnel; forming a first set by the association vectors of all the persons, and obtaining unit vectors in the principal component direction of the first set to obtain a first space; transforming the vectors in the first set to a first space to obtain a second set; setting target personnel, and obtaining the replaceability of the target personnel in the principal component direction according to the direction similarity between the vector corresponding to the target personnel in the second set and the unit vector in the principal component direction and the projection length of each vector of the second set in the unit vector in the principal component direction; obtaining the replaceability of the target personnel according to the replaceability of the target personnel in the principal component direction and the motion state distribution characteristics of the target personnel and the equipment; and carrying out safety production area personnel management according to the replaceability of the target personnel.

8. The system of claim 7, wherein the data analysis module is further configured to:

obtaining a unit vector of a space where the first set is located to obtain a second space, and transforming the unit vector in the principal component direction of the first space to the second space, wherein a transformation coefficient corresponding to the unit vector of the second space is a weight coefficient in the principal component direction; and weighting and superposing the motion state distribution characteristics of the target person and all the devices by using the weight coefficients in the principal component directions, and obtaining the substitutability of the target person according to the substitutability of the target person in all the principal component directions and the superposition result.

9. The system of claim 7, wherein the data acquisition module is further configured to:

and acquiring the motion state distribution of the personnel at each position of the equipment neighborhood when the personnel arrive at the equipment every time, and weighting and superposing the motion state distribution to obtain the motion state distribution characteristics of the personnel and the equipment.

10. The system according to claim 9, wherein the step of weighted superposition of motion state distributions comprises:

and determining a weight coefficient according to the variation of the danger degree of the equipment when the personnel arrive at the equipment every time, and performing weighted superposition on the motion state distribution according to the weight coefficient to obtain the motion state distribution characteristics of the personnel and the equipment.

Background

In the production process of enterprises, the safety is of great importance. In the safety production area, there are many production or storage facilities, such as chemical reaction and storage for chemicals, or high temperature and high pressure supply for chemical reaction, etc., which are inevitably faulty or abnormal during operation, thereby causing a dangerous accident and causing large-scale casualties in severe cases, and therefore, it is very important to manage personnel in the safety production area.

The workers who participate in the production need to operate, patrol, overhaul and the like the equipment. Personnel are important participants in maintaining regional safety and discovering and handling hazards in a timely manner. An unreasonable personnel management method can cause unreasonable allocation of human resources, which is not beneficial to efficient production on one hand and safe maintenance of a production area on the other hand.

Disclosure of Invention

In order to solve the problems, the invention provides a safety production area personnel management method based on big data analysis, which comprises the following steps:

acquiring a motion state sequence of equipment when a person arrives at a safe production area and a danger degree change sequence of the equipment when the person arrives beside the equipment; analyzing the association relation between the motion state sequence and the danger degree change sequence to obtain the association characteristics of the personnel and the equipment;

obtaining the motion state distribution characteristics of the personnel and the equipment according to the motion states of the personnel at the positions in the neighborhood of the equipment:

acquiring the association characteristics of a person and all equipment to obtain an association vector of the person; forming a first set by the association vectors of all the persons, and obtaining unit vectors in the principal component direction of the first set to obtain a first space;

transforming the vectors in the first set to a first space to obtain a second set; setting target personnel, and obtaining the replaceability of the target personnel in the principal component direction according to the direction similarity between the vector corresponding to the target personnel in the second set and the unit vector in the principal component direction and the projection length of each vector of the second set in the unit vector in the principal component direction; obtaining the replaceability of the target personnel according to the replaceability of the target personnel in the principal component direction and the motion state distribution characteristics of the target personnel and the equipment; and carrying out safety production area personnel management according to the replaceability of the target personnel.

Preferably, a unit vector of a space where the first set is located is obtained to obtain a second space, the unit vector in the principal component direction of the first space is transformed into the second space, and a transformation coefficient corresponding to the unit vector of the second space is a weight coefficient in the principal component direction; and weighting and superposing the motion state distribution characteristics of the target person and all the devices by using the weight coefficients in the principal component directions, and obtaining the substitutability of the target person according to the substitutability of the target person in all the principal component directions and the superposition result.

Preferably, when the person arrives at the equipment each time, the movement state of the person at each position in the neighborhood of the equipment is obtained to obtain the movement state distribution, and the movement state distribution is weighted and superposed to obtain the movement state distribution characteristics of the person and the equipment.

Preferably, a weight coefficient is determined according to the variation of the risk degree of the equipment when the personnel arrive at the equipment every time, and the motion state distribution characteristics of the personnel and the equipment are obtained by performing weighted superposition on the motion state distribution according to the weight coefficient.

Preferably, the time series convolution network is used for analyzing the association relation between the motion state sequence and the danger degree change sequence to obtain the association grade as the association characteristic of the personnel and the equipment.

Preferably, the association relationship between the motion state sequence of the equipment and the risk degree change sequence of the equipment when the personnel arrive at the safe production area each time is obtained, and the association characteristics of the personnel and the equipment are obtained according to the average value of all the association relationships.

The invention also provides a safety production area personnel management system based on big data analysis, which comprises:

the data acquisition module is used for acquiring a motion state sequence of equipment when personnel arrive at a safe production area and a danger degree change sequence of the equipment when the personnel arrive beside the equipment;

the system comprises a relation characteristic acquisition module, a risk degree change sequence acquisition module and a risk degree change sequence acquisition module, wherein the relation characteristic acquisition module is used for analyzing the relation between the motion state sequence and the risk degree change sequence to obtain the relation characteristics of personnel and equipment; obtaining the motion state distribution characteristics of the personnel and the equipment according to the motion state of the personnel at each position in the neighborhood of the equipment;

the data analysis module is used for acquiring the association characteristics of the personnel and all the equipment to obtain the association vector of the personnel; forming a first set by the association vectors of all the persons, and obtaining unit vectors in the principal component direction of the first set to obtain a first space; transforming the vectors in the first set to a first space to obtain a second set; setting target personnel, and obtaining the replaceability of the target personnel in the principal component direction according to the direction similarity between the vector corresponding to the target personnel in the second set and the unit vector in the principal component direction and the projection length of each vector of the second set in the unit vector in the principal component direction; obtaining the replaceability of the target personnel according to the replaceability of the target personnel in the principal component direction and the motion state distribution characteristics of the target personnel and the equipment; and carrying out safety production area personnel management according to the replaceability of the target personnel.

The invention has the following beneficial effects:

according to the method, the association characteristics and the state distribution characteristics of the personnel and the equipment are obtained through the movement track of the personnel in the safe production area and the danger degree of the equipment, and the relation between the personnel and the equipment is accurately reflected. According to the invention, the degree of substitution of personnel and the degree of substitution of equipment are obtained through the correlation characteristics and the state distribution characteristics of personnel and equipment, so that personnel management in a safe production area is carried out, the human resources are reasonably utilized, unnecessary personnel are prevented from being present in the safe production area, meanwhile, the safety risk of the production area due to the lack of necessary production personnel is avoided, the human resources are reasonably and efficiently used, and meanwhile, the safety of the production area is ensured.

Drawings

FIG. 1 is a process flow diagram.

Detailed Description

The invention provides a management method of a safe production area, aiming at timely finding out unnecessary production personnel or missing personnel, increasing or decreasing the personnel in the production area, reasonably utilizing human resources, avoiding the occurrence of the unnecessary personnel in the safe production area and avoiding the increase of the safety risk of the production area due to the lack of the necessary production personnel. In the prior art, personnel management in a safe production area is mostly concentrated on personnel activity monitoring, the problem of fundamental operator management and allocation is not solved, and the complex relation between personnel and equipment is ignored, so that the personnel management is unreasonable. The invention is described in further detail below with reference to the figures and specific examples.

The first embodiment is as follows:

the embodiment provides a safety production area personnel management method based on big data analysis, and a flow chart of the method is shown in fig. 1.

The invention obtains the replaceability of the target person based on the association characteristics of the person and the equipment and the motion state distribution characteristics.

Firstly, acquiring a motion state sequence of equipment when a person arrives at a safe production area and a danger degree change sequence of the equipment when the person arrives beside the equipment; and acquiring the motion state distribution of the personnel at each position of the equipment neighborhood when the personnel arrive at the equipment every time.

In the embodiment, a key point detection network is used to obtain key points (points where feet of people touch the ground) of each person in the image from the image, the key points are used as position coordinates of the person in the image, and the position coordinates of the person are mapped to a ground coordinate system through affine transformation.

Each device has status data indicating the status of the device, such as the pressure, temperature, chemical remaining amount, etc. inside the device, and when the device is in a normal state, the data is within a normal range, and when the data is beyond the normal range, the device is considered to be out of order or abnormal. And acquiring state data of each device exceeding the normal range, inputting the data exceeding the normal range into the fully-connected neural network (if the data does not exceed the normal range, the size of the data exceeding the normal range is considered to be 0, and the labels of the data sets are artificially labeled), and outputting the risk degree of the device, wherein the risk degree is used for representing the probability of the device being abnormal or failed, or the risk degree is generated. The degree of risk is classified into ten levels in this embodiment: 0.0, 0.1, … …, 0.9.

And acquiring the position of each person at different time according to the historical data, and acquiring the position of each person at different time in production. When a person is present beside a device (centered on the device and with a radius within a circular area worth of a predetermined value), the risk level a of the device before entering the device is recorded, as well as the risk level b of the person leaving the device. The risk degree of the equipment varies by an amount ofThe value indicates a change in the risk level of the equipment when a person approaches the equipment, and if the value is large, the person is considered to be able to reduce the risk level of the equipment, for example, to perform maintenance on the equipment, to adjust the operating parameters of the equipment, and the like.

Specifically, regarding the motion state sequence of the device when the person arrives at the safety production area and the method for obtaining the motion state of the person at each position in the neighborhood of the device, the first implementation method is obtained by a traditional speed statistical method, and the embodiment also provides a special filtering method for obtaining the motion state of the person.

For a person and a certain device, the person can appear beside the device for multiple times, and the danger degree variable quantity of the device is acquired when the ith person is beside the device. Obtaining the nearest time when the person arrives at the deviceMotion trajectory in time unit (in the present embodiment)One second is taken as a time unit), and the motion trail refers to a change sequence of the position coordinates of the person along with time. The sequence is subjected to special mean filtering (the reason that a conventional mean filtering method is not used is that sequence elements are vectors and direct mean taking is not appropriate), and the method adopted in the embodiment is as follows: obtain a length ofWhen the window is slid every time, a Gaussian hot spot is generated by taking each position coordinate in the window as a center, each position on the Gaussian hot spot corresponds to a heat value, a plurality of Gaussian hot spots can be obtained in one window, and forgetting coefficients of the Gaussian hot spots are superposed.

For these heightsThe method for superposing the forgetting coefficients of the ephemeral hot spots comprises the following steps: the plurality of hot spots in the window are arranged in sequence according to the time sequence. Assuming that any position in any production area is p, the heat value of the position p on the hot spots is pIf the position p is outside a certain hot spot, the heat value of the position p on the hot spot is 0. The final heat value of the position p is obtained by using the forgetting coefficientWhereinIs a forgetting coefficient, is a hyper-parameter, in this embodimentAnd is and。smaller heat values, which indicate that these heat spots are generated sequentially over time without accumulating or superimposing higher heat values at the position p, indicate faster movement of the person, thisThe final heat superposition result at the position p is called, the final heat superposition results at all positions are obtained, the sum x of all the results which are greater than a preset threshold (the value is 0.2 in the embodiment) is obtained, and the smaller the value is, the higher heat value is not accumulated in the heat spots generated by the person at all positions in the movement process, namely the person moves faster. I.e. one x is obtained at a time by sliding the window,the value representing the motion state as the result of filtering corresponding to the position of the center of the window (or the time at which the center of the window is located) at one sliding window (i.e., the result corresponds to one position and one time), and the larger the value, the faster the person moves at the position, and the more the person focuses on the device. In this embodiment, the window step size is 2, and sliding is performed to obtain a filtering result.

And filtering the sequence, wherein the filtering result after each sliding window corresponds to one position and one moment. Therefore, two filtering results are obtained after the whole motion track is subjected to multiple sliding windows, one result is that the motion states of the personnel at different times, namely the sequence of the motion speed changing along with the time is called the motion state sequence of the equipment(ii) a Another result is the movement state of the persons at different positions, namely the movement speed of the persons at different positions; if the person is not present at a certain position, the motion state at the position is 0, that is, the motion states of all the persons at the position are called the person motion state distribution. In addition, the nearest of the person when arriving next to the device is obtainedThe time-varying sequence of the risk level of the unit equipment in a time is called the risk level varying sequence of the equipment。

And then, analyzing the association relation between the motion state sequence and the danger degree change sequence by utilizing a time sequence convolution network to obtain an association grade as the association characteristic of the personnel and the equipment. Acquiring the association relationship between the motion state sequence of the equipment when the personnel arrives at the safe production area and the danger degree change sequence of the equipment when the personnel arrives beside the equipment, and acquiring the association characteristics of the personnel and the equipment according to the average value of all the association relationships. And determining a weight coefficient according to the variation of the danger degree of the equipment when the personnel arrive at the equipment every time, and performing weighted superposition on the motion state distribution according to the weight coefficient to obtain the motion state distribution characteristics of the personnel and the equipment.

To pairAndin thatWhen (1), (b)When i takes different values, the sequence length is the same, and the same principle is adoptedAlso) willAndinputting a scalar into a TCN networkIn this embodiment, the value range of the value isThe value indicates the relevance rank of the two sequences, i.e. the relevance rank is divided into 11 ranks, the larger the person is concerned about the change of the danger level of the equipment.，The mean value of (a) represents the person-to-device association characteristic,a larger means that the degree of danger of the device is more relevant to the state of motion of the person, indicating that the person is more relevant or concerned with the device. The input to the TCN network is two sequences, so that the data set for the TCN network can be easily generated in a large number of computer simulations.

Distribution of person movement state（Indicating for a person and a device the distribution of the person's state of motion when the person is present next to the device for the ith time),and then, overlapping, wherein the overlapping method comprises the following steps: set at any positionThe personnel state ofThen, thenThe superposition result is，When all values are takenThe result of the value set of (a) is a superposition result, wherein the weight coefficient is(the amount of change in the degree of risk of the apparatus is set to). The superposition result indicates whether the personnel pay attention to the change of the danger degree of the equipment at the position of the personnel, and indicates the working state or working condition of the personnel during production work. The larger the value of each position on the superposition result is, the better the working state of the staff is, and the more beneficial the reduction of the danger degree of the equipment is. The superposition result is called the motion state distribution characteristics of the personnel and the equipment.

Finally, obtaining the association characteristics of the personnel and all the devices to obtain the association vector of the personnel; forming a first set by the association vectors of all the persons, and obtaining unit vectors in the principal component direction of the first set to obtain a first space; transforming the vectors in the first set to a first space to obtain a second set; and obtaining a unit vector of the space where the first set is located to obtain a second space, and transforming the unit vector in the principal component direction of the first space to the second space. And setting target persons, and obtaining the replaceability of the target persons in the principal component direction according to the direction similarity between the vector corresponding to the target persons in the second set and the unit vector in the principal component direction and the projection length of each vector of the second set in the principal component direction. The transformation coefficient corresponding to the second space unit vector is the weight coefficient of the principal component direction; weighting and superposing the motion state distribution characteristics of the target person and all equipment by using the weight coefficients in the principal component directions, and obtaining the substitutability of the target person according to the substitutability of the target person in all principal component directions and superposition results; and carrying out safety production area personnel management according to the replaceability of the target personnel.

In this embodiment, it is assumed that there are R devices and Q persons. For a person, acquiring the association features of the person and all R devices, wherein the R association features form an association vector of the person, and the association vector set of Q persons is a first set。

For the first setPrincipal component analysis of the obtainedThe principal component directions of the vectors represent the main distribution directions of the associated vectors, namely, the principal component directions represent the directions of R principal components in the direction with the maximum projection variance of the associated vectors. Obtaining a unit vector set of the principal component directions of the first set to obtain a first space, wherein the unit vector set of the space is a unit vector set. The first setTransforming into the first space to obtain a second setThe space dimension of the first set is R, R base vectors of the R-dimension space form a second space, and the unit vector set of the space is(each basis vector therein corresponds to a device).

The target person q's substitutability in the direction of the r-th principal component is obtained as follows:

in the formula (I), the compound is shown in the specification,representing row vectorsAnd unit row vectorInner product of (2), representing a vectorIn thatThe longer the projection length, the more important the correlation vector of the person q is in the r-th principal component direction.To representWherein Q associated vectors are inThe sum of the lengths of the projections on the optical disk,indicating in addition to the relevance vectorOther association vectors than those described above areThe sum of the projected lengths of (a) and (b).

Indicating in addition to the relevance vectorOther association vectors than those described above areSum of projection lengths and associated vectorIn thatThe larger the ratio, the more the correlation vector isIn thatThe smaller the ratio of the upper projection length is, the more the upper projection length can be replaced by other association vectors; the smaller the ratio, the more relevant the vector isIn thatThe larger the ratio of the up-projection length, the less likely it is to be replaced by other association vectors.

The larger the value is, the more the association vector is specifiedIn thatThe smaller the ratio of the upper projection length is, and the other association vectors are inIf the sum of the lengths of the upper projections is large, the correlation vector is interpretedThe less important.

That is, the direction similarity between the vector corresponding to the target person and the unit vector in the principal component direction, the cosine similarity is adopted in this embodiment to represent the vectorAnd vectorCosine ofSimilarity, the larger the vectorAnd vectorThe smaller the included angle. Larger values indicate greater attentionThe size of (2).

In general terms, the number of active devices,the larger the person q is, the larger the replaceability in the direction of the r-th principal component. Vector of the r-th principal componentIs a linear weighted summation of the basis vectors in V2, i.e. each vector in V2 represents the characteristics of only one device, i.e.Representing features of multiple devices when combined, if a person is associated with a vectorThe longer the projected length is, the higher the relevance or concern of the person to the plurality of devices at the same time, i.e. the person has to be responsible for the plurality of devices at the same time, and to improve the service of maintenance, operation, etc. for the plurality of devices.

A weight coefficient is obtainedSatisfy the following requirements，To representThe xth basis vector (the xth basis vector represents the xth device); when in useThe characteristic of the combination of the immediate component direction and the xth device,the larger, the descriptionThe tighter the bond with the xth device, ifVery small, even 0, indicatesNot integrated with the xth device.

The alternatives for target person q are then:,representing the distribution of the motion state of the qth individual and the xth device,represents a weighted sum of the motion state distribution characteristics of the qth person and all devices,the larger the vector, the larger the x-th device and the r-th principal component direction unit vectorThe tighter the bond.

As described above, the motion states of the person q and the device x are dividedCloth featuresIndicating whether the person q is in the location and responding positively to changes in the risk level of the equipment x, the sum of the distribution characteristics of the person q and all the equipment is indicative of whether the person q is in the location and responding positively to changes in the risk level of all the equipment x. Then the weighted sum of the motion state distribution characteristics of the qth person and all devicesThat is, whether the person q is at the position of allThe change in the degree of risk of the incorporated device responds positively.

Because of the fact thatA distribution characteristic of the motion states of the qth individual and the xth device (the distribution characteristic of the motion states can be regarded as an image and a thermodynamic diagram representing the distribution of the states), each position on the distribution corresponding to a state value,the result is a fused feature distribution, and each position on the fused feature distribution also corresponds to a state value.The accumulated summation of the state values of each position on the distribution is shown, and the larger the result is, the more the personnel can actively correspond or deal with the fault of the equipment when the personnel are in most positions, and the like, the smaller the personnel replaceability is.

The larger the two are, the larger the person is in the r-th principal component directionThe greater the replaceability. Since the principal component r characterizes the equipment that the person q needs to take care of or pay attention to.The person q is instead characterized in handling the devices, i.e. how important the person q is to the devices.The larger the representation, the more the person q can be replaced and the less important the person q.

Similarly, the present embodiment can obtain the replaceability of the device r，Andthe calculation method is the same as that of the method: the method comprises the following steps: for one device r, acquiring the association features of all the persons and the device to form an association vector of the device, and acquiring the substitutability of each device in each principal component direction by using principal component analysis, thereby acquiring the substitutability of the device r through the distribution features of the persons and the device. The larger the device r can be replaced, the less important the device is in production, and for the device, in order to increase the safety of the device, the additional hands are needed to pay attention to or overhaul or operate the device to avoid the failure and danger.

And carrying out safety production area personnel management according to the personnel replaceability and the equipment replaceability. When the person replaceability is greater than the threshold, the person can be left out of the production area, avoiding danger to the person. When the replaceability of the equipment is larger than the threshold value, more human hands are added and distributed to the equipment, so that the equipment is prevented from being dangerous and affecting safe production.

The above embodiments are merely preferred embodiments of the present invention, which should not be construed as limiting the present invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention should be included in the scope of the present invention.

Example two:

the embodiment provides a safety production area personnel management system based on big data analysis, and the system comprises:

the data acquisition module is used for acquiring a motion state sequence of equipment when personnel arrive at a safe production area and a danger degree change sequence of the equipment when the personnel arrive beside the equipment;

the system comprises a relation characteristic acquisition module, a risk degree change sequence acquisition module and a risk degree change sequence acquisition module, wherein the relation characteristic acquisition module is used for analyzing the relation between the motion state sequence and the risk degree change sequence to obtain the relation characteristics of personnel and equipment; obtaining the motion state distribution characteristics of the personnel and the equipment according to the motion state of the personnel at each position in the neighborhood of the equipment;

the data analysis module is used for acquiring the association characteristics of the personnel and all the equipment to obtain the association vector of the personnel; forming a first set by the association vectors of all the persons, and obtaining unit vectors in the principal component direction of the first set to obtain a first space; transforming the vectors in the first set to a first space to obtain a second set; setting target personnel, and obtaining the replaceability of the target personnel in the principal component direction according to the direction similarity between the vector corresponding to the target personnel in the second set and the unit vector in the principal component direction and the projection length of each vector of the second set in the unit vector in the principal component direction; obtaining the replaceability of the target personnel according to the replaceability of the target personnel in the principal component direction and the motion state distribution characteristics of the target personnel and the equipment; and carrying out safety production area personnel management according to the replaceability of the target personnel.

The above embodiments are merely preferred embodiments of the present invention, which should not be construed as limiting the present invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention should be included in the scope of the present invention.