A distribution network non-power-off operation risk early warning method and device and computer equipment

Abstract

The application discloses a distribution network non-power-off operation risk early warning method and device and computer equipment, and relates to the technical field of distribution network operation safety. The method is to firstly correlate the position and the electric field intensity according to the Kinect bone data collected by a Kinect device in real time and the electric field intensity collected by a MEMS electric field sensor in real time, obtain the electric field intensity measured values at multiple measuring points, then inversely obtain the electric field three-dimensional distribution of the distribution network non-power-off operation site, and finally determine the corresponding current electric field intensity according to the electric field three-dimensional distribution and the corresponding current position of each human body joint point, and determine whether there is a human body electric shock risk, if yes, trigger the early warning action, so that the operation personnel can be reminded to correct the mistake in time, further personnel electric shock safety accidents can be avoided, the safety of the operation personnel and equipment can be ensured, and the practical application and popularization are facilitated.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network operation safety technology, specifically relating to a risk early warning method, device and computer equipment for power distribution network live-line operation. Background Technology

[0002] Live-line work on distribution networks is a service in the power industry that ensures the normal operation of the power grid (i.e., troubleshooting power faults without interrupting power supply to ensure normal electricity use for production and daily life). The specific workflow for live-line work on distribution networks includes: personnel reporting requirements -- on-site inspection -- arranging maintenance plans -- work scheduling -- on-site work -- work evaluation, etc.

[0003] When working on live power lines, the low voltage and electromagnetic field strength of the power grid, the small spatial distance between conductors, and the large number of electrical devices all pose significant challenges. Workers are easily exposed to electrical equipment within their working area, making operation difficult. To ensure the safety of personnel and equipment, multiple safety measures are currently implemented, consisting of primary and auxiliary insulating tools, along with methods such as using insulating rods and insulating gloves for on-site operations.

[0004] However, in actual field operations, due to factors such as varying personnel quality, non-standard operating procedures, and / or weak self-protection awareness, workers may unknowingly perform dangerous actions, resulting in their bodies getting too close to live conductors and causing electric shock accidents. Therefore, how to provide early warnings and reminders of dangers during live-line work is a topic that urgently needs to be studied by those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method, device, computer equipment, and computer-readable storage medium for risk warning during live-line work in power distribution networks, in order to solve the problem of electric shock accidents that may occur due to workers' unintentional dangerous actions during live-line work in power distribution networks.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] Firstly, a risk early warning method for live-line work in distribution networks is provided, including:

[0008] The system acquires Kinect skeletal data collected in real time from the worker by a Kinect device and electric field strength collected in real time by at least two MEMS electric field sensors. The Kinect device is deployed at the power distribution network uninterrupted operation site, ensuring its field of view covers the worker's work area. The Kinect skeletal data package contains the three-dimensional coordinates of multiple human joints of the worker. These multiple human joints include any combination of head joints, neck joints, spinal shoulder joints, spinal midline joints, spinal base joints, left shoulder joints, left elbow joints, left wrist joints, left hand joints, left hip joints, left knee joints, left ankle joints, left foot joints, right shoulder joints, right elbow joints, right wrist joints, right hand joints, right hip joints, right knee joints, right ankle joints, and right foot joints. Each of the at least two MEMS electric field sensors corresponds one-to-one with at least two of the multiple human joints, and the MEMS electric field sensors are attached to the corresponding human body parts at those joints.

[0009] For each of the at least two human joints, the electric field intensity collected by the corresponding MEMS electric field sensor at any historical moment is taken as the measured value of the electric field intensity at the corresponding historical position at any historical moment.

[0010] Based on the measured values ​​of electric field strength at multiple measurement points, the three-dimensional distribution of electric field at the power distribution network live-line operation site is obtained by inversion. The multiple measurement points include all historical locations of the at least two human joint points in the most recent historical period.

[0011] For each of the multiple human joints, the current electric field strength is determined based on the three-dimensional distribution of the electric field and the corresponding current position.

[0012] For each of the multiple human body joints, the system determines whether there is a risk of electric shock based on the corresponding current electric field strength. If so, an early warning action is triggered.

[0013] Based on the above-mentioned invention, an early warning scheme is provided that can automatically identify the risk of electric shock to workers during live-line power distribution work. First, the scheme correlates the position with the electric field strength by using Kinect skeletal data collected in real-time from the worker and MEMS electric field sensors to obtain measured electric field strength values ​​at multiple measurement points. Then, it inversely calculates the three-dimensional distribution of the electric field at the live-line power distribution work site. Finally, for each joint point on the worker, based on the three-dimensional electric field distribution and the corresponding current position, it determines the current electric field strength and assesses whether there is a risk of electric shock. If so, an early warning action is triggered. This alerts workers to correct errors promptly, preventing further electric shock accidents and ensuring the safety of workers and equipment. This approach is suitable for practical application and widespread adoption.

[0014] In one possible design, the at least two human joints include any combination of the head joint, the left hand joint, the left foot joint, the right hand joint, and the right foot joint.

[0015] In one possible design, based on the measured electric field strength values ​​at multiple measurement points, the three-dimensional electric field distribution at the power distribution network live-line work site is inverted, including:

[0016] Based on the measured electric field strength values ​​at multiple measurement points, the three-dimensional electric field distribution at the power distribution network live-line operation site is obtained by inversion using the Kriging interpolation method. The multiple measurement points include all historical locations of the at least two human joints in the most recent historical period.

[0017] In one possible design, based on the measured electric field strength values ​​at multiple measurement points, the three-dimensional electric field distribution at the power distribution network live-line work site is obtained by inversion using the Kriging interpolation method, including:

[0018] Based on the measured values ​​of electric field strength at multiple measured points, the semivariance of electric field strength for each pair of measured points is calculated to obtain the semivariance values ​​of electric field strength for multiple pairs of measured points. In addition, the distance values ​​of multiple pairs of measured points are calculated based on the known coordinates of the multiple measured points. The multiple measured points include all historical positions of the at least two human joints in the most recent historical period.

[0019] Determine the lag distance h1 and the longest distance H among the distance values ​​of the multiple pairs of measured points. max Where h1 represents a positive number, H max This represents a positive number greater than h1;

[0020] Based on the hysteresis h1, the first interval (0, H) maxThe interval is divided into the following first subintervals: (0,h1],(h1,2*h1],…,((k-1)*h1,k*h1],…,((K-1)*h1,H max ], where K = Ceiling(H max / h1), Ceiling() represents the floor function, and k represents a positive integer less than k;

[0021] Based on the distance values ​​of the multiple pairs of measured points and the attribution relationship of the multiple first sub-intervals, the multiple pairs of measured points are divided into multiple first groups that correspond one-to-one with the multiple first sub-intervals;

[0022] Based on the electric field strength semivariance and distance values ​​of the multiple pairs of measured points, the average electric field strength semivariance and average distance of each of the multiple first groups are calculated.

[0023] Based on the average semivariogram of electric field intensity and the average distance of each first group, the model coefficients of multiple experimental variation function models are fitted. The average semivariogram of electric field intensity is used as the experimental variation function value during the fitting process, and the average distance is used as the distance to the point to be estimated as a regional variable during the fitting process.

[0024] Based on the measured values ​​of electric field intensity at the multiple measured points, error analysis is performed using the model coefficients of the multiple experimental variogram models to obtain the model quality evaluation index value of each experimental variogram model in the multiple experimental variogram models.

[0025] Based on the model quality evaluation index values ​​of each experimental variation function model, the optimal experimental variation function model that best meets the preset conditions for model optimization is determined from the multiple experimental variation function models.

[0026] Based on the known coordinates of the plurality of measured points and the known coordinates of the target measured point in the power distribution network live-line operation site, determine m measured points located around the target measured point from the plurality of measured points, where m represents a positive integer greater than 2;

[0027] Based on the known coordinates of the target measurement point and the known coordinates of the m measured points, the distance from the target measurement point to each of the m measured points is calculated. This distance is then used as a regional variable to represent the distance to the point to be estimated and substituted into the optimal experimental variation function model. Finally, the model coefficients of the optimal experimental variation function model are applied to calculate the experimental variation function values ​​between the target measurement point and each of the m measured points.

[0028] Based on the electric field strength semivariance values ​​of each pair of measured points among the m measured points and the experimental variation function values ​​of the target measuring point and each of the m measured points, the following ordinary Kriging equations are established:

[0029]

[0030] In the formula, i and j represent positive integers, and λ i γ(x) represents the weight coefficient to be solved for the i-th measured point among the m measured points. i ,x j ) represents the semivariance of the electric field intensity corresponding to the i-th measured point and the j-th measured point among the m measured points, u represents the Lagrange multiplier factor to be solved, and γ(x) represents the semivariance of the electric field intensity corresponding to the i-th measured point and the j-th measured point among the m measured points. i (x0) represents the experimental variation function value between the target measurement point and the i-th measured point;

[0031] Solving the ordinary Kriging equations yields m weighting coefficients that correspond one-to-one with the m measured points.

[0032] Based on the measured electric field strength values ​​at the m measured points, the estimated electric field strength Z(x0) at the target measuring point is calculated using the following formula:

[0033]

[0034] In the formula, Z(x) i ) represents the measured value of the electric field strength at the i-th measured point;

[0035] The estimated electric field intensity of all target measuring points in the entire distribution network live-line work site will be used as the three-dimensional distribution of the electric field in the distribution network live-line work site.

[0036] In a possible design, determining the hysteresis h1 includes the following steps S3031~S3034:

[0037] S3031. In the interval (0, H) max Select a value from the [value] as the current value of the hysteresis distance h1, and then execute step S3032, where H max This represents the longest distance value among the distance values ​​of the multiple pairs of measured points;

[0038] S3032. For each experimental variation function model in multiple experimental variation function models, based on the current value of the lag distance h1 and the measured values ​​of the electric field strength at the multiple measured points, cross-validation is used to obtain the corresponding estimated values ​​of the electric field strength at the multiple measured points, and then step S3033 is executed.

[0039] S3033. For each experimental variation function model, calculate the corresponding model quality evaluation index value based on the measured electric field strength values ​​at the multiple measured points and the corresponding estimated electric field strength values ​​at the multiple measured points, and then execute step S3034.

[0040] S3034. Determine whether the model quality assessment index values ​​of each experimental variation function model meet the preset iteration stopping condition. If yes, then determine the current value of the lag distance h1 as the final value; otherwise, in the interval (0, H... max Select a new value as the current value of the hysteresis distance h1, and then execute step S3032.

[0041] In one possible design, for each experimental variogram model among multiple experimental variogram models, based on the current value of the hysteresis distance h1 and the measured electric field strength values ​​at the multiple measured points, cross-validation is used to obtain the corresponding estimated electric field strength values ​​at the multiple measured points, including:

[0042] For each target measured point among the plurality of measured points, all other measured points among the plurality of measured points are determined as corresponding plurality of reference measured points;

[0043] For a specific experimental variation function model among multiple experimental variation function models and a specific target measured point among multiple measured points, based on the current value of the hysteresis distance h1, the model parameters of the specific experimental variation function model, and the measured electric field strength values ​​of multiple reference measured points of the specific target measured point, the corresponding estimated electric field strength value is calculated according to the following steps:

[0044] Determine the longest distance value among multiple pairs of reference measured points. Among them, the multiple pairs of reference measured points refer to a point pair among the multiple reference measured points of a certain target measured point;

[0045] Based on the current value of the lag distance h1, the second interval It is divided into the following multiple second sub-intervals: in, Ceiling() is a function that rounds up the value. Indicates less than Positive integers;

[0046] Based on the distance values ​​of the multiple pairs of reference measured points and the attribution relationship of the multiple second sub-intervals, the multiple pairs of reference measured points are divided into multiple second groups that correspond one-to-one with the multiple second sub-intervals;

[0047] Based on the electric field strength semivariance and distance values ​​of the multiple pairs of reference measured points, the average electric field strength semivariance and average distance of each of the multiple second groups are calculated.

[0048] Based on the average semivariogram of electric field intensity and the average distance of each second group, the model coefficients of a certain experimental variation function model are fitted, wherein the average semivariogram of electric field intensity is used as the experimental variation function value during the fitting process, and the average distance is used as the distance to the point to be estimated as a regionalization variable during the fitting process.

[0049] Based on the known coordinates of multiple reference measured points of a certain target measured point and the known coordinates of the target measured point, the distance value from the target measured point to each of the multiple reference measured points is calculated. This distance value is then used as a regional variable to the distance to the point to be estimated and substituted into the experimental variation function model. Then, the model parameters of the experimental variation function model are applied to calculate the experimental variation function value between the target measured point and each of the multiple reference measured points of the target measured point.

[0050] Based on the electric field strength semivariance values ​​of each pair of reference measured points among multiple reference measured points of a certain target measured point and the experimental variation function values ​​of each reference measured point among multiple reference measured points of the certain target measured point, the ordinary Kriging equation system is established and solved to obtain multiple reference weight coefficients that correspond one-to-one with the multiple reference measured points of the certain target measured point.

[0051] The estimated value of the electric field strength at the target measurement point is calculated based on the measured values ​​of the electric field strength at multiple reference measurement points and the multiple reference weighting coefficients.

[0052] In one possible design, for a specific human joint among the plurality of human joints, the presence of a risk of electric shock is determined based on the corresponding current electric field strength, including:

[0053] For a specific human joint among the plurality of human joints, the corresponding current voltage U is calculated according to the current electric field strength E, using the following formula:

[0054] U = E × d

[0055] In the formula, d represents the insulation thickness of the insulating protective component, which is used to provide electrical protection to the human body part corresponding to a certain human joint point;

[0056] Determine whether the current voltage U reaches the product of the insulation voltage of the insulating protective component and a preset coefficient. If it does, then it is determined that there is a risk of electric shock to the human body. The preset coefficient has a value range of [0.6, 0.9].

[0057] Secondly, a risk early warning device for live-line work in distribution networks is provided, which includes a data acquisition module, an electric field strength correlation module, an electric field distribution inversion module, a current field strength determination module, and a risk early warning triggering module that are connected in sequence via communication.

[0058] The data acquisition module is used to acquire Kinect skeleton data collected in real time from the worker by the Kinect device and electric field strength collected in real time by at least two MEMS electric field sensors. The Kinect device is deployed at the power distribution network uninterrupted operation site and the lens field of view covers the worker's work area. The Kinect skeleton data package contains the three-dimensional coordinates of multiple human joints of the worker. The multiple human joints include any combination of head joints, neck joints, spine shoulder joints, spine mid-section joints, spine base joints, left shoulder joints, left elbow joints, left wrist joints, left hand joints, left hip joints, left knee joints, left ankle joints, left foot joints, right shoulder joints, right elbow joints, right wrist joints, right hand joints, right hip joints, right knee joints, right ankle joints, and right foot joints. The at least two MEMS electric field sensors correspond one-to-one with at least two of the multiple human joints, and the MEMS electric field sensors are attached to the corresponding human body parts at the joints.

[0059] The electric field intensity correlation module is used to take the electric field intensity collected by the corresponding MEMS electric field sensor at any historical moment as the measured value of the electric field intensity at the historical location of each of the at least two human joints at any historical moment.

[0060] The electric field distribution inversion module is used to invert the three-dimensional electric field distribution at the power distribution network live-line operation site based on the measured values ​​of electric field strength at multiple measured points. The multiple measured points include all the historical locations of the at least two human joints in the most recent historical period.

[0061] The current field strength determination module is also communicatively connected to the data acquisition module, and is used to determine the corresponding current electric field strength for each of the multiple human joints based on the three-dimensional distribution of the electric field and the corresponding current position.

[0062] The risk warning triggering module is used to determine whether there is a risk of electric shock to each of the multiple human joint points based on the corresponding current electric field strength. If so, a warning action is triggered.

[0063] Thirdly, the present invention provides a computer device comprising a memory, a processor, and a transceiver connected in sequence for communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the distribution network uninterrupted operation risk warning method as described in the first aspect or any possible design in the first aspect.

[0064] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, perform the distribution network live-line operation risk warning method as described in the first aspect or any possible design in the first aspect.

[0065] Fifthly, the present invention provides a computer program product containing instructions that, when the instructions are executed on a computer, cause the computer to perform the distribution network live-line operation risk warning method as described in the first aspect or any possible design in the first aspect.

[0066] The beneficial effects of the above scheme are:

[0067] (1) This invention creatively provides an early warning scheme that can automatically identify the risk of electric shock to workers during power distribution network live-line work. First, the position and electric field strength are correlated based on the Kinect skeleton data collected in real time by the Kinect device and the electric field strength collected in real time by the MEMS electric field sensor to obtain the measured values ​​of electric field strength at multiple measurement points. Then, the three-dimensional distribution of electric field at the power distribution network live-line work site is obtained by inversion. Finally, for each human joint point, the corresponding current electric field strength is determined based on the three-dimensional distribution of electric field and the corresponding current position, and it is judged whether there is a risk of electric shock to the human body. If so, an early warning action is triggered. This can remind workers to correct their mistakes in time, avoid further electric shock accidents, ensure the safety of workers and equipment, and facilitate practical application and promotion.

[0068] (2) It can also quickly and efficiently obtain the three-dimensional electric field distribution results that meet the usage requirements and accuracy requirements with only a small amount of measured data from actual measurement points, greatly shortening the required time, and thus having certain theoretical significance and high engineering practical value.

[0069] (3) The appropriate hysteresis h1 that can be obtained by cross-validation and iteration can be automatically determined so as to facilitate the quick and efficient acquisition of the final three-dimensional electric field distribution results. Attached Figure Description

[0070] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0071] Figure 1 This is a flowchart illustrating the risk warning method for live-line work in power distribution networks provided in this application embodiment.

[0072] Figure 2 Example diagram of 25 human joints in Kinect skeletal data provided for embodiments of this application.

[0073] Figure 3 This is a schematic diagram of the structure of the power distribution network live-line operation risk early warning device provided in the embodiments of this application.

[0074] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0075] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0076] It should be understood that although the terms "first" and "second", etc., may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another. For example, the first object may be referred to as the second object, and similarly, the second object may be referred to as the first object, without departing from the scope of the exemplary embodiments of the invention.

[0077] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, or A and B exist simultaneously. Another example is A, B and / or C, which can mean that any one of A, B, and C or any combination thereof exists. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone or A and B exist simultaneously. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0078] Example:

[0079] like Figure 1 As shown, the power distribution network live-line operation risk early warning method provided in the first aspect of this embodiment can be executed, but is not limited to, by a computer device with certain computing resources and which is communicatively connected to a Kinect device (a type of existing depth-sensing camera developed by Microsoft; "Kinect" is a neologism created by adding "Connection" to "Kinetics"), a MEMS (Micro-Electro-Mechanical System) electric field sensor, and an early warning device. For example, it can be executed by an electronic device such as a field server, a personal computer (PC, referring to a multi-purpose computer of a size, price, and performance suitable for personal use; desktop computers, laptops, mini-laptops, tablets, and ultrabooks all belong to personal computers), a smartphone, a personal digital assistant (PDA), or a wearable device. Figure 1 As shown, the method for early warning of risks during power outage operations in the distribution network may include, but is not limited to, the following steps S1 to S5.

[0080] S1. Acquire Kinect skeleton data collected in real time from the worker by the Kinect device and electric field strength collected in real time by at least two MEMS electric field sensors. The Kinect device is deployed at the power distribution network uninterrupted operation site and the lens field of view covers the worker's work area. The Kinect skeleton data package contains the three-dimensional coordinates of multiple human joints of the worker. The multiple human joints include, but are not limited to, any combination of head joints, neck joints, spine shoulder joints, spine mid-section joints, spine base joints, left shoulder joints, left elbow joints, left wrist joints, left hand joints, left hip joints, left knee joints, left ankle joints, left foot joints, right shoulder joints, right elbow joints, right wrist joints, right hand joints, right hip joints, right knee joints, right ankle joints, and right foot joints. The at least two MEMS electric field sensors correspond one-to-one with at least two of the multiple human joints, and the MEMS electric field sensors are attached to the corresponding human body parts at the joints.

[0081] In step S1, the Kinect device is preferably the second-generation Kinect V2 device, which, compared to the first-generation Kinect device, features an improved 4-array microphone (zero-balanced), new active infrared detection, and 1080P high-definition video. It can also utilize improved skeletal tracking and recognition technology to separate the foreground and background of the depth image, and then track 6 complete bones and 25 joints (e.g., based on depth image features) using skeletal joint point tracking. Figure 2As shown, the joint points are: Head (1), Neck (2), Spine Shoulder (3), Spine Mid (4), Spine Base (5), Shoulder Left (6), Elbow Left (7), Wrist Left (8), Hand Left (9), Thumb Left (10), Hand Tip Left (11), Hip Left (12), Knee Left (13), and Ankle Left (14). The Kinect skeletal data is generated by tracking the following joints: left foot (15), right shoulder (16), right elbow (17), right wrist (18), right hand (19), right thumb (20), right fingertip (21), right hip (22), right knee (23), right ankle (24), and right foot (25). The data is then transmitted at approximately 30fps, resulting in more accurate and stable posture tracking. Since the Kinect device tracks these 25 joints, the resulting Kinect skeletal data inevitably includes the three-dimensional coordinates of these joints.

[0082] In step S1, the MEMS electric field sensor is an existing device used to measure electric field strength. For example, a silicon-based MEMS electric field measurement sensor developed by a research team at the Vienna University of Technology (TU Wien) is used. This sensor does not interfere with the electric field measurement process. Its measurement principle is as follows: a mass is suspended on an elastic element, which is fixed to a conductive frame. When placed in an electric field, electrostatic induction generates a force on the mass. The mass will displace under this force. The displacement is then measured using optical principles to obtain the electric field strength that is positively correlated with the displacement. This silicon-based MEMS electric field measurement sensor can measure electric field strength from low frequencies to up to 1 kHz, and can also reliably measure weak electric fields of less than 200 volts / meter.

[0083] In step S1, considering that the human head, hands, and feet are human body parts with a large range of motion, to ensure that more measured values ​​of electric field strength can be obtained subsequently, preferably, the at least two human joint points include, but are not limited to, any combination of the head joint point, the left hand joint point, the left foot joint point, the right hand joint point, and the right foot joint point. That is, the MEMS electric field sensor can be attached to the human head corresponding to the head joint point (specifically, it can be installed in an insulated helmet), to the left / right hand corresponding to the left / right hand joint point (specifically, it can be installed in an insulated glove), and to the left / right foot corresponding to the left / right foot joint point (specifically, it can be installed in an insulated shoe). In addition, the Kinect skeleton data and the electric field strength need to be collected synchronously, and the local device can receive and acquire the Kinect skeleton data and the electric field strength through conventional wireless / wired communication methods.

[0084] S2. For each of the at least two human joints, the electric field intensity collected by the corresponding MEMS electric field sensor at any historical moment is taken as the measured value of the electric field intensity at the corresponding historical position at any historical moment.

[0085] In step S2, for example, for a left hand joint, if the corresponding historical position at a certain historical moment is a three-dimensional coordinate X, then the electric field intensity collected by the corresponding MEMS electric field sensor attached to the left hand at that historical moment can be used as the measured value of the electric field intensity at the three-dimensional coordinate X. Furthermore, if there are multiple measured values ​​of the electric field intensity at different historical moments for a certain historical position, then the average value, a conventional weighted average, or the measured value of the electric field intensity at the most recent historical moment is taken as the final measured value of the electric field intensity.

[0086] S3. Based on the measured values ​​of electric field strength at multiple measured points, the three-dimensional distribution of electric field at the power distribution network live-line operation site is obtained by inversion, wherein the multiple measured points include all historical locations of the at least two human joints in the most recent historical period.

[0087] In step S3, since the electric field strength decreases nonlinearly as the distance from the measuring point to the charged body decreases, the measured electric field strength of the global space can be obtained by using the measured electric field strength of the local space (i.e., the measured electric field strength of the multiple measuring points) based on this characteristic of the electric field space, thereby obtaining the three-dimensional distribution of the electric field at the power distribution network live-line operation site. Since the electric field strength at a point is related to and can be derived from the electric field strength of its surrounding points, the three-dimensional electric field distribution at the live-line work site can be obtained by inverting the measured electric field strength values ​​at multiple measured points using Kriging interpolation (also known as the Kriging method, a spatial interpolation method proposed by South African engineer Krige DG; its basic assumption is that the attribute value of a point is related to the attribute values ​​of its surrounding points and can be derived from the attribute values ​​of its surrounding points; it is an optimal, unbiased estimation method with strong spatial correlation, using variograms / variance functions as calculation tools and combined with structural analysis). The multiple measured points include all historical positions of the at least two human joints in the most recent historical period. Considering that the electric field strength at the work site may exhibit time-varying characteristics due to the movement of charged bodies, all historical positions need to be subject to a time limit: within the most recent historical period (e.g., within the last 3 minutes) to ensure that the obtained three-dimensional electric field distribution is up-to-date. Furthermore, the three-dimensional distribution of the electric field can be displayed in the form of isosurfaces, and can be updated by inversion as the measured points and / or measured values ​​of the electric field intensity are updated.

[0088] S4. For each of the multiple human joints, determine the current electric field strength based on the three-dimensional distribution of the electric field and its current position.

[0089] S5. For each of the multiple human joint points, determine whether there is a risk of electric shock to the human body based on the corresponding current electric field strength. If so, trigger an early warning action.

[0090] In step S5, specifically, for a specific human joint among the plurality of human joints, the presence of a risk of electric shock is determined based on the corresponding current electric field strength, including but not limited to the following steps S51 to S52: S51. For a specific human joint among the plurality of human joints, the corresponding current voltage U is calculated based on the corresponding current electric field strength E according to the following formula:

[0091] U = E × d

[0092] In the formula, d represents the insulation thickness of the insulating protective component, which is used to provide electrical protection to the human body part corresponding to a certain human joint point; S52. Determine whether the current voltage U reaches the result of multiplying the insulation voltage of the insulating protective component by a preset coefficient. If it does, it is determined that there is a risk of electric shock to the human body, wherein the preset coefficient is in the range of [0.6, 0.9]. Specific examples of the aforementioned insulating protective components include: an insulating helmet corresponding to the head joint point, insulating gloves corresponding to the left / right hand joint points, insulating shoes corresponding to the left / right foot joint points, and insulating clothing corresponding to other joint points. Furthermore, the triggering of the warning action may include, but is not limited to, issuing a voice prompt message corresponding to the human joint point with a risk of electric shock through a warning reminder device such as a loudspeaker. For example, when a risk of electric shock is found at the left elbow joint point, a voice prompt message such as "Caution! Risk of electric shock to the left elbow! Please avoid" can be issued to remind workers to correct their mistakes in time, avoid further electric shock accidents, and ensure the safety of workers and equipment.

[0093] Therefore, based on the risk warning method for live-line work in power distribution networks described in steps S1 to S5 above, a warning scheme is provided that can automatically identify the risk of electric shock to workers during live-line work in power distribution networks. First, the location and electric field strength are correlated based on the Kinect skeletal data collected in real time by the Kinect device and the electric field strength collected in real time by the MEMS electric field sensor to obtain the measured electric field strength values ​​at multiple measurement points. Then, the three-dimensional distribution of the electric field at the live-line work site is obtained from this. Finally, for each human joint, the corresponding current electric field strength is determined based on the three-dimensional distribution of the electric field and the corresponding current location, and it is judged whether there is a risk of electric shock. If so, a warning action is triggered. This can remind workers to correct their mistakes in time, avoid further electric shock accidents, ensure the safety of workers and equipment, and facilitate practical application and promotion.

[0094] Based on the technical solution of the first aspect mentioned above, this embodiment also provides a possible design for how to obtain the three-dimensional distribution of the electric field by inversion using the Kriging interpolation method. That is, based on the measured values ​​of the electric field intensity at multiple measured points, the three-dimensional distribution of the electric field at the power supply uninterrupted operation site of the distribution network is obtained by inversion using the Kriging interpolation method, including but not limited to the following steps S301 to S314.

[0095] S301. Based on the measured values ​​of electric field strength at multiple measured points, calculate the semivariance of electric field strength for each pair of measured points to obtain the semivariance values ​​of electric field strength for multiple pairs of measured points. Also, based on the known coordinates of the multiple measured points, calculate the distance values ​​of the multiple pairs of measured points. The multiple measured points include all historical positions of the at least two human joints in the most recent historical period.

[0096] In step S301, the electric field strength semivariance represents half the square of the difference between the measured electric field strength values ​​at two measuring points, i.e., the electric field strength semivariance value between measuring point A and measuring point B. Among them, Z A Z represents the measured value of the electric field strength at measuring point A. B This represents the measured value of the electric field strength at measurement point B. Since the semivariance of the electric field strength between any two points is related to the distance between those two points, it is also necessary to calculate the distance values ​​between the multiple pairs of measured points.

[0097] Before step S301, considering that this embodiment is based on the ordinary kriging method for estimation, and the ordinary kriging method requires the data to follow a normal distribution, preferably, before calculating the semivariance of the electric field intensity for each pair of measured points based on the measured electric field intensity values ​​at multiple measured points, the method further includes, but is not limited to: determining whether the measured electric field intensity values ​​at the multiple measured points follow a normal distribution; if not, performing a power transformation or logarithmic transformation on the measured electric field intensity values ​​at the multiple measured points to obtain new measured electric field intensity values ​​at the multiple measured points that follow a normal distribution. The specific method of the aforementioned determination is an existing conventional method, such as based on the condition of following a normal distribution (i.e., if the random variable X follows a mathematical expectation of μ and a variance of σ). 2 The normal distribution is denoted as N(μ,σ). 2 The judgment is made by performing a power transformation. The specific formula for the aforementioned power transformation process is: The specific formula for the aforementioned logarithmic transformation is as follows: Where Z(x) represents the measured value of the electric field strength before processing. This represents the newly measured value of the electric field strength obtained after processing, and η represents a preset positive coefficient.

[0098] S302. Determine the hysteresis distance h1 and the longest distance value H among the distance values ​​of the multiple pairs of measured points. max Where h1 represents a positive number, H max This represents a positive number greater than h1.

[0099] In step S302, the hysteresis distance is an academic term in the Kriging method, and it can be determined manually, randomly, or automatically.

[0100] S303. Based on the hysteresis distance h1, the first interval (0, H) is... max The interval is divided into the following first subintervals: (0,h1],(h1,2*h1],…,((k-1)*h1,k*h1],…,((K-1)*h1,H max ], where K = Ceiling(H max / h1), Ceiling() represents the floor function, and k represents a positive integer less than k.

[0101] In step S303, for example, if the lag distance h1 is 1m, the longest distance value H max If the distance is 10 meters, then the first interval can be divided into 10 (i.e., K=10) first sub-intervals.

[0102] S304. Based on the distance values ​​of the multiple pairs of measured points and the attribution relationship of the multiple first sub-intervals, divide the multiple pairs of measured points into multiple first groups that correspond one-to-one with the multiple first sub-intervals.

[0103] In step S304, for example, if the distance between a pair of measured points is 6.8, then the pair of measured points can be assigned to a first group corresponding to a first sub-interval (6,7], and so on. Furthermore, for non-last groups among the multiple first groups, there may be cases where no pair of measured points is assigned.

[0104] S305. Based on the electric field strength semivariance and distance values ​​of the multiple pairs of measured points, calculate the average electric field strength semivariance and average distance of each of the multiple first groups.

[0105] S306. Based on the average semivariogram of electric field intensity and the average distance of each first group, the model coefficients of multiple experimental variation function models are fitted, wherein the average semivariogram of electric field intensity is used as the experimental variation function value during the fitting process, and the average distance is used as the distance to the point to be estimated as a regional variable during the fitting process.

[0106] In step S306, the experimental variation function model is an important research tool of the Kriging method. Specifically, the multiple experimental variation function models include, but are not limited to, any combination of the following models (A) to (C):

[0107] (A) Spherical model, expressed as:

[0108] (B) Exponential model, expressed as:

[0109] (C) Gaussian model, expressed as:

[0110] In the above expression, γ(h) represents the experimental variogram value, h represents the distance from the regionalized variable to the point to be estimated, C0 represents the sill value as the first model coefficient, C represents the camber (i.e., the maximum value of the spatial variation of the regionalized variable) as the second model coefficient, and a represents the range (i.e., the range within which the regionalized variable has correlation) as the third model coefficient. Therefore, the model coefficients to be fitted include the sill value C0, the camber C, and the range a. The specific fitting method can be, but is not limited to, the least squares method. Furthermore, for the non-last group among the multiple first groups that has no measured point pair to which it belongs, considering that the average semivariogram of the corresponding electric field intensity and the average distance are both zero, they have no fitting value and need to be skipped in the fitting process.

[0111] S307. Based on the measured values ​​of electric field strength at the multiple measured points, error analysis is performed using the model coefficients of the multiple experimental variogram models to obtain the model quality evaluation index value of each experimental variogram model in the multiple experimental variogram models.

[0112] In step S307, the model quality evaluation index values ​​include, but are not limited to, the mean error value, the root mean square error value, the standardized root mean square error value, and / or the mean standard error value. The calculation formulas for the aforementioned index values ​​are as follows:

[0113] Average error value

[0114] Root mean square error

[0115] Standardized root mean square error

[0116] Mean standard error

[0117] In the above formula, N represents the total number of measured points, n represents a positive integer, and Z(x) n () represents the measured value of the electric field strength at the nth measurement point. δ represents the estimated electric field strength at the nth measured point. 2 This represents the square root of the variance. Since obtaining the model quality assessment index value requires an estimated electric field strength value, specifically, based on the measured electric field strength values ​​at the multiple measured points, error analysis is performed using the model coefficients of the multiple experimental variogram models to obtain the model quality assessment index value for each experimental variogram model, including but not limited to the following steps S3071~S3072.

[0118] S3071. For each experimental variation function model in the plurality of experimental variation function models, based on the measured values ​​of electric field intensity at the plurality of measured points and the corresponding model coefficients, the corresponding estimated values ​​of electric field intensity at the plurality of measured points are obtained by cross-validation.

[0119] In step S3071, the specific idea of ​​the cross-validation method is as follows: first, remove one of the multiple measured points, and then use the measured electric field strength values ​​of the remaining measured points to calculate the estimated electric field strength value of the removed point. Repeat this operation until the estimated electric field strength values ​​of all measured points are obtained. Specifically, for each experimental variogram model in the multiple experimental variogram models, based on the measured electric field strength values ​​of the multiple measured points and the corresponding model coefficients, the cross-validation method is used to obtain the corresponding estimated electric field strength values ​​of the multiple measured points, including but not limited to the following steps S30711 to S30712.

[0120] S30711. For each target measured point (i.e., a removed measured point) among the plurality of measured points, all other measured points (i.e., the remaining measured points) among the plurality of measured points are determined as the corresponding plurality of reference measured points.

[0121] S30712. For a certain experimental variation function model among multiple experimental variation function models and a certain target measured point among multiple measured points, the corresponding electric field strength estimate is calculated according to the model parameters of the certain experimental variation function model and the measured electric field strength values ​​of multiple reference measured points of the target measured point, following the steps S307121 to S307123.

[0122] S307121. Based on the known coordinates of multiple reference measured points of the target measured point and the known coordinates of the target measured point, calculate the distance value from the target measured point to each of the multiple reference measured points of the target measured point, and substitute this distance value as the distance to the point to be estimated into the experimental variation function model. Then, apply the model parameters of the experimental variation function model to calculate the experimental variation function value between the target measured point and each of the multiple reference measured points of the target measured point.

[0123] S307122. Based on the electric field strength semivariance values ​​of each pair of reference measured points among the multiple reference measured points of the target measured point and the experimental variation function values ​​of each reference measured point among the multiple reference measured points of the target measured point, establish and solve the ordinary Kriging equation system to obtain multiple reference weight coefficients that correspond one-to-one with the multiple reference measured points of the target measured point.

[0124] In step S307122, the specific establishment and solution process of the ordinary Kriging equations can be derived by referring to the subsequent steps S311 to S312, and will not be repeated here.

[0125] S307123. Based on the measured values ​​of the electric field strength at multiple reference measured points of the target measured point and the multiple reference weighting coefficients, the estimated value of the electric field strength at the target measured point is calculated.

[0126] In step S307123, the specific calculation formula can be derived by referring to the subsequent step S313, and will not be repeated here.

[0127] S3072. For each experimental variation function model, the corresponding model quality evaluation index value is calculated based on the measured electric field strength values ​​at the multiple measured points and the corresponding estimated electric field strength values ​​at the multiple measured points.

[0128] S308. Based on the model quality evaluation index values ​​of each experimental variation function model, determine the optimal experimental variation function model that best meets the preset conditions for model optimization from among the multiple experimental variation function models.

[0129] In step S308, specifically, the preferred preset conditions for the model include, but are not limited to, an average error value close to 0, a standardized root mean square error value close to 1, and / or a root mean square error value close to the average standard error value. The more the aforementioned preferred preset conditions for the model are met, the better the quality of the corresponding model.

[0130] S309. Based on the known coordinates of the plurality of measured points and the known coordinates of the target measured point in the power distribution network live-line operation site, determine m measured points located around the target measured point from the plurality of measured points, where m represents a positive integer greater than 2.

[0131] In step S309, the target measurement point can be a real measurement point or a non-real measurement point (i.e., a measurement point that needs to be valued). The area surrounding the target measurement point can specifically refer to a circular region centered on the target measurement point with a radius of a specific value. This specific value can be adjusted appropriately based on the search results of the real measurement point. For example, the radius value can be increased when m is too small, and decreased when m is too large.

[0132] S310. Based on the known coordinates of the target measuring point and the known coordinates of the m measured points, calculate the distance from the target measuring point to each of the m measured points, and substitute this distance as the distance to the point to be estimated into the optimal experimental variation function model. Then, apply the model coefficients of the optimal experimental variation function model to calculate the experimental variation function values ​​between the target measuring point and each of the m measured points.

[0133] S311. Based on the semivariance values ​​of the electric field intensity of each pair of measured points among the m measured points and the experimental variation function values ​​of the target measuring point and each of the m measured points, the following ordinary Kriging equations are established:

[0134]

[0135] In the formula, i and j represent positive integers, and λ i γ(x) represents the weight coefficient to be solved for the i-th measured point among the m measured points. i ,x j ) represents the semivariance of the electric field intensity corresponding to the i-th measured point and the j-th measured point among the m measured points, u represents the Lagrange multiplier factor to be solved, and γ(x) represents the semivariance of the electric field intensity corresponding to the i-th measured point and the j-th measured point among the m measured points. i x0) represents the experimental variation function value between the target measurement point and the i-th measured point.

[0136] In step S311, the ordinary Kriging equation system is a combination of equations (m+1) established based on the Kriging interpolation method, wherein... This is the unbiased estimation condition for the Kriging interpolation method.

[0137] S312. Solve the ordinary Kriging equations to obtain m weighting coefficients that correspond one-to-one with the m measured points.

[0138] In step S312, since there are only m+1 unknowns in the ordinary Kriging equations, the m weight coefficients corresponding to the m measured points can be obtained by conventional equation-solving methods.

[0139] S313. Based on the measured electric field strength values ​​at the m measured points, the estimated electric field strength Z(x0) at the target measuring point is calculated using the following formula:

[0140]

[0141] In the formula, Z(x) i ) represents the measured value of the electric field strength at the i-th measured point.

[0142] In step S313, since the measured electric field strength value can be used as the attribute value of a known point, and the electric field strength can be regarded as multiple realizations of a random field, the electric field strength estimation result at the power supply interruption site of the distribution network can be obtained by estimating / interpolating using the above formula.

[0143] S314. The estimated electric field intensity of all target measuring points in the entire distribution network live-line work site shall be used as the three-dimensional electric field distribution of the distribution network live-line work site.

[0144] Therefore, based on the aforementioned possible design one, the three-dimensional electric field distribution results that meet the usage requirements and accuracy requirements can be obtained quickly and efficiently using only a small amount of measured data from actual measurement points, greatly shortening the required time, and thus having certain theoretical significance and high engineering practical value.

[0145] Based on the aforementioned possible design one technical solution, this embodiment also provides a possible design two for how to automatically determine the appropriate hysteresis distance h1, that is, to determine the hysteresis distance h1, including but not limited to the following steps S3031 to S3034.

[0146] S3031. In the interval (0, H) max Select a value from the [value] as the current value of the hysteresis distance h1, and then execute step S3032, where H max This represents the longest distance value among the distance values ​​of the multiple pairs of measured points.

[0147] In step S3031, the specific selection method may be, but is not limited to, random selection.

[0148] S3032. For each experimental variation function model in multiple experimental variation function models, based on the current value of the lag distance h1 and the measured values ​​of the electric field strength at the multiple measured points, cross-validation is used to obtain the corresponding estimated values ​​of the electric field strength at the multiple measured points, and then step S3033 is executed.

[0149] In step S3032, the specific approach of the cross-validation method is similar to that in step S3071. Specifically, for each experimental variogram model in the multiple experimental variogram models, based on the current value of the lag distance h1 and the measured values ​​of the electric field strength at the multiple measured points, the cross-validation method is used to obtain the corresponding estimated values ​​of the electric field strength at the multiple measured points, including but not limited to the following steps S30321 to S30322:

[0150] S30321. For each target measured point among the plurality of measured points, all other measured points among the plurality of measured points are determined as corresponding plurality of reference measured points.

[0151] S30322. For a certain experimental variation function model among multiple experimental variation function models and a certain target measured point among multiple measured points, the corresponding estimated value of electric field intensity is calculated according to the current value of the lag distance h1, the model parameters of the certain experimental variation function model, and the measured values ​​of electric field intensity at multiple reference measured points of the target measured point, following the steps S303221 to S303228.

[0152] S303221. Determine the longest distance value among multiple pairs of reference measured points. The multiple pairs of reference measured points refer to a pair of points among the multiple reference measured points of a certain target measured point.

[0153] S303222. Based on the current value of the hysteresis distance h1, the second interval... It is divided into the following multiple second sub-intervals:

[0154] in,

[0155] Ceiling() is a function that rounds up the value. Indicates less than Positive integers.

[0156] S303223. Based on the distance values ​​of the multiple pairs of reference measured points and the attribution relationship of the multiple second sub-intervals, divide the multiple pairs of reference measured points into multiple second groups that correspond one-to-one with the multiple second sub-intervals.

[0157] S303224. Based on the electric field strength semivariance and distance values ​​of the multiple pairs of reference measured points, calculate the average electric field strength semivariance and average distance of each of the multiple second groups.

[0158] S303225. Based on the average semivariogram of electric field intensity and the average distance of each of the second groups, the model coefficients of the experimental variation function model are fitted, wherein the average semivariogram of electric field intensity is used as the experimental variation function value during the fitting process, and the average distance is used as the distance to the point to be estimated as a regional variable during the fitting process.

[0159] S303226. Based on the known coordinates of multiple reference measured points of the target measured point and the known coordinates of the target measured point, calculate the distance value from the target measured point to each of the multiple reference measured points of the target measured point, and substitute this distance value as the distance to the point to be estimated into the experimental variation function model. Then, apply the model parameters of the experimental variation function model to calculate the experimental variation function value between the target measured point and each of the multiple reference measured points of the target measured point.

[0160] S303227. Based on the electric field strength semivariance values ​​of each pair of reference measured points among the multiple reference measured points of the target measured point and the experimental variation function values ​​of each reference measured point among the multiple reference measured points of the target measured point, establish and solve the ordinary Kriging equation system to obtain multiple reference weight coefficients that correspond one-to-one with the multiple reference measured points of the target measured point.

[0161] S303228. Based on the measured values ​​of the electric field strength at multiple reference measured points of the target measured point and the multiple reference weighting coefficients, the estimated value of the electric field strength at the target measured point is calculated.

[0162] The specific details of steps S303221 to S303228 above can be derived by referring to similar steps in the first aspect, and will not be repeated here.

[0163] S3033. For each experimental variation function model, calculate the corresponding model quality evaluation index value based on the measured electric field strength values ​​at the multiple measured points and the corresponding estimated electric field strength values ​​at the multiple measured points, and then execute step S3034.

[0164] S3034. Determine whether the model quality assessment index values ​​of each experimental variation function model meet the preset iteration stopping condition. If yes, then determine the current value of the lag distance h1 as the final value; otherwise, in the interval (0, H... max Select a new value as the current value of the hysteresis distance h1, and then execute step S3032.

[0165] In step S3034, the iteration stopping condition can be set based on, but is not limited to, some index thresholds. For example, it may include an average error value less than a preset average error threshold and / or a standardized root mean square error value greater than a preset standardized root mean square error threshold, etc., to measure whether a relatively ideal experimental variation function model has been obtained. Furthermore, the specific method of reselection can also be, but is not limited to, random selection.

[0166] Therefore, based on the aforementioned possible design two, the appropriate hysteresis h1 that can obtain a more ideal model can be automatically determined through cross-validation and iteration, so as to further facilitate the quick and efficient acquisition of the final three-dimensional electric field distribution results.

[0167] like Figure 3 As shown, the second aspect of this embodiment provides a virtual device for implementing the risk warning method for live-line work in distribution networks as described in the first aspect or any possible design in the first aspect. The device includes a data acquisition module, an electric field strength correlation module, an electric field distribution inversion module, a current field strength determination module, and a risk warning triggering module that are connected in sequence.

[0168] The data acquisition module is used to acquire Kinect skeleton data collected in real time from the worker by the Kinect device and electric field strength collected in real time by at least two MEMS electric field sensors. The Kinect device is deployed at the power distribution network uninterrupted operation site and the lens field of view covers the worker's work area. The Kinect skeleton data package contains the three-dimensional coordinates of multiple human joints of the worker. The multiple human joints include any combination of head joints, neck joints, spine shoulder joints, spine mid-section joints, spine base joints, left shoulder joints, left elbow joints, left wrist joints, left hand joints, left hip joints, left knee joints, left ankle joints, left foot joints, right shoulder joints, right elbow joints, right wrist joints, right hand joints, right hip joints, right knee joints, right ankle joints, and right foot joints. The at least two MEMS electric field sensors correspond one-to-one with at least two of the multiple human joints, and the MEMS electric field sensors are attached to the corresponding human body parts at the joints.

[0169] The electric field intensity correlation module is used to take the electric field intensity collected by the corresponding MEMS electric field sensor at any historical moment as the measured value of the electric field intensity at the historical location of each of the at least two human joints at any historical moment.

[0170] The electric field distribution inversion module is used to invert the three-dimensional electric field distribution at the power distribution network live-line operation site based on the measured values ​​of electric field strength at multiple measured points. The multiple measured points include all the historical locations of the at least two human joints in the most recent historical period.

[0171] The current field strength determination module is also communicatively connected to the data acquisition module, and is used to determine the corresponding current electric field strength for each of the multiple human joints based on the three-dimensional distribution of the electric field and the corresponding current position.

[0172] The risk warning triggering module is used to determine whether there is a risk of electric shock to each of the multiple human joint points based on the corresponding current electric field strength. If so, a warning action is triggered.

[0173] The working process, working details and technical effects of the aforementioned device provided in the second aspect of this embodiment can be found in the first aspect or any possible design in the first aspect, which describes the risk warning method for live-line work in distribution networks, and will not be repeated here.

[0174] like Figure 4 As shown, the third aspect of this embodiment provides a computer device for executing the distribution network live-line work risk warning method as described in the first aspect or any possible design in the first aspect. The device includes a memory, a processor, and a transceiver connected in sequence. The memory stores a computer program, the transceiver sends and receives messages, and the processor reads the computer program to execute the distribution network live-line work risk warning method as described in the first aspect or any possible design in the first aspect. Specifically, the memory may include, but is not limited to, random-access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or first-in-last-out (FILO) memory, etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. Furthermore, the computer device may also include, but is not limited to, a power module, a display screen, and other necessary components.

[0175] The working process, working details and technical effects of the aforementioned computer equipment provided in the third aspect of this embodiment can be found in the first aspect or any possible design in the first aspect, which describes the risk warning method for live-line work in distribution networks, and will not be repeated here.

[0176] The fourth aspect of this embodiment provides a computer-readable storage medium storing instructions comprising the distribution network live-line work risk warning method as described in the first aspect or any possible design within the first aspect. Specifically, the computer-readable storage medium stores instructions that, when executed on a computer, perform the distribution network live-line work risk warning method as described in the first aspect or any possible design within the first aspect. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or Memory Sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0177] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be found in the distribution network live-line operation risk warning method as described in the first aspect or any possible design in the first aspect, and will not be repeated here.

[0178] This fifth aspect of the embodiment provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the power distribution network live-line operation risk early warning method as described in the first aspect or any possible design within the first aspect. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0179] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A power distribution network operation risk early warning method, characterized in that, The method comprises: acquiring Kinect skeleton data of a worker collected by a Kinect device in real time and electric field intensity collected by at least two MEMS electric field sensors in real time, wherein the Kinect device is arranged at a live-line distribution work site and has a lens field of view covering a work area of the worker, the Kinect skeleton data contains three-dimensional coordinates of multiple human body joints of the worker, and the at least two MEMS electric field sensors correspond to at least two human body joints of the multiple human body joints one by one, and the MEMS electric field sensors are attached to human body parts corresponding to the joints; for each human body joint of the at least two human body joints, taking electric field intensity collected by a corresponding MEMS electric field sensor at any historical time as an actual measurement value of electric field intensity at a corresponding historical position at the any historical time; According to the measured values of the electric field intensity at multiple measurement points, the three-dimensional distribution of the electric field at the distribution network live-line working site is obtained by inverse calculation based on the Kriging interpolation method, specifically including: according to the measured values of the electric field intensity at multiple measurement points, the semi-variance of the electric field intensity of each pair of measurement points in the multiple measurement points is calculated respectively to obtain the semi-variance values of the electric field intensity of multiple pairs of measurement points, and the distance values of the multiple pairs of measurement points are also calculated according to the known coordinates of the multiple measurement points, wherein the multiple measurement points contain all historical positions of the at least two human body joints in the nearest historical period; determine the lag distance and the longest distance value in the distance values of the multiple pairs of measurement points , wherein is a positive number, is a positive number greater than ; according to the lag distance , the first interval is divided into multiple first subintervals as follows: , wherein , is a rounding up function, is a positive integer less than ; according to the relationship between the distance values of the multiple pairs of measurement points and the multiple first subintervals, the multiple pairs of measurement points are divided into multiple first groups corresponding to the multiple first subintervals one by one; according to the semi-variance values and distance values of the multiple pairs of measurement points, the semi-variance average value and distance average value of each first group in the multiple first groups are calculated; according to the semi-variance average value and distance average value of each first group, the model coefficients of multiple experimental variogram models are fitted, wherein the semi-variance average value is used as the experimental variogram value in the fitting process, and the distance average value is used as the distance from the regionalized variable to the to-be-estimated point in the fitting process; according to the measured values of the electric field intensity of the multiple measurement points, error analysis is performed by applying the model coefficients of the multiple experimental variogram models to obtain the model quality evaluation index value of each experimental variogram model in the multiple experimental variogram models; according to the model quality evaluation index value of each experimental variogram model, the optimal experimental variogram model that best meets the model optimization preset condition is determined from the multiple experimental variogram models; according to the known coordinates of the multiple measurement points and the known coordinates of a target measurement point in the distribution network live-line working site, the measurement points located around the target measurement point are determined from the multiple measurement points, wherein is a positive integer greater than 2; according to the known coordinates of the target measurement point and the known coordinates of the measurement points, the distance from the target measurement point to the distance values of each of the measured points, and the distance values are taken as the distance of the regionalized variable to the target point into the optimal experimental variogram model, and then the model coefficients of the optimal experimental variogram model are applied to calculate the distance between the target point and the experimental variogram values of each of the measured points; according to the electrical field intensity semi-variance values of each pair of the measured points, and the distance between the target point and the experimental variogram values of each of the measured points, the following ordinary Kriging equation group is established: In the formula, and respectively represent positive integers, represent the weight coefficients corresponding to the first measured point among the measured points and to be solved, represent the weight coefficients corresponding to the first measured point among the measured points and to be solved, represent the electric field intensity semi-variance values corresponding to the first measured point among the measured points and to be solved, represent the experimental variogram values of the target measured point and the first measured point; the ordinary Kriging equation group is solved to obtain weight coefficients corresponding to the measured points one by one; and the electric field intensity estimated value of the target measured point is calculated according to the electric field intensity measured values of the measured points in the following formula In the formula, represents the measured value of the electric field intensity of the first measured point; the estimated value of the electric field intensity of all the target measured points in the whole domain of the distribution network live working site is taken as the three-dimensional distribution of the electric field of the distribution network live working site. for each human body joint of the multiple human body joints, determining a corresponding current electric field intensity according to the three-dimensional distribution of the electric field and a corresponding current position; for each human body joint of the multiple human body joints, judging whether there is a human body electric shock risk according to the corresponding current electric field intensity, and if so, triggering a warning action.

2. The power distribution network non-power-off operation risk early warning method according to claim 1, characterized in that, The multiple human body joints include any combination of a head joint, a neck joint, a spine shoulder joint, a spine middle joint, a spine base joint, a left shoulder joint, a left elbow joint, a left wrist joint, a left hand joint, a left hip joint, a left knee joint, a left ankle joint, a left foot joint, a right shoulder joint, a right elbow joint, a right wrist joint, a right hand joint, a right hip joint, a right knee joint, a right ankle joint, and a right foot joint.

3. The power distribution network non-power-off operation risk early warning method according to claim 2, characterized in that, The at least two human body joints include any combination of the head joint, the left hand joint, the left foot joint, the right hand joint, and the right foot joint.

4. The power distribution network non-power-off operation risk early warning method according to claim 1, characterized in that, Determining a hysteresis distance including the steps S3031-S3034: S3031. In the interval a value is selected as the current value of the hysteresis distance and step S3032 is performed, in which represents the longest distance value among the distance values of the pairs of measured points. S3032. For each experimental variance function model in the plurality of experimental variance function models, according to the current value of the lag distance and the measured values of the electric field intensity of the plurality of measured points, the corresponding estimated values of the electric field intensity of the plurality of measured points are obtained by using cross-validation, and then step S3033 is performed. S3032. For each experimental variance function model in the plurality of experimental variance function models, according to the current value of the lag distance and the measured values of the electric field intensity of the plurality of measured points, the corresponding estimated values of the electric field intensity of the plurality of measured points are obtained by using cross-validation, and then step S3033 is performed. S3033. For each experimental variogram model, a model quality evaluation index value is calculated according to the electric field intensity actual measurement values of the multiple measurement points and corresponding electric field intensity estimated values of the multiple measurement points, and then step S3034 is performed; S3034. Determine whether the model quality evaluation index value of each experimental variance function model satisfies the preset iteration stop condition, if yes, determine the current value of the lag distance as the final value, otherwise, reselect a value within the interval as the current value of the lag distance , and then execute step S3032.

5. The power distribution network non-power-off operation risk early warning method according to claim 4, characterized in that, For each experimental variation function model in the plurality of experimental variation function models, according to the current value of the lag distance and the measured values of the electric field intensity of the plurality of measured points, the corresponding and the estimated values of the electric field intensity of the plurality of measured points are obtained in a cross-validation manner, comprising: for each target measurement point in the multiple measurement points, all other measurement points in the multiple measurement points are determined as corresponding multiple reference measurement points; For a specific experimental variation function model among multiple experimental variation function models and a specific target measured point among the multiple measured points, based on the lag distance... The current value of the electric field, the model parameters of the experimental variation function model, and the measured electric field strength values ​​of multiple reference measurement points of the target measurement point are used to calculate the corresponding estimated electric field strength value according to the following steps: determining a longest distance value among distance values of pairs of reference measured points wherein the pairs of reference measured points refer to all pairs of points among the multiple reference measured points of the certain target measured point According to the current value of the hysteresis distance , the second interval is divided into a plurality of second sub-intervals as follows: wherein, , denotes a ceiling function, denotes a positive integer smaller than . according to the distance values of the multiple pairs of reference measurement points and the attribution relationships of the multiple second subintervals, the multiple pairs of reference measurement points are divided into multiple second groups corresponding to the multiple second subintervals one by one; according to the electric field intensity semi-variance values and the distance values of the multiple pairs of reference measurement points, electric field intensity semi-variance average values and distance average values of each second group in the multiple second groups are calculated; according to the electric field intensity semi-variance average values and the distance average values of the multiple second groups, model coefficients of the certain experimental variogram model are fitted, wherein the electric field intensity semi-variance average values are used as experimental variogram values in the fitting process, and the distance average values are used as distances of regionalized variables to the to-be-estimated points in the fitting process. According to the known coordinates of the multiple reference measured points of the certain target measured point and the known coordinates of the certain target measured point, the distance value of the certain target measured point to each reference measured point of the multiple reference measured points of the certain target measured point is calculated, and the distance value is taken as the distance of the regionalized variable to the to-be-estimated point into the certain experimental variogram model, and then the model parameters of the certain experimental variogram model are applied to calculate the experimental variogram value of the certain target measured point and each reference measured point of the multiple reference measured points of the certain target measured point; According to the electrical field intensity semi-variance value of each pair of reference measured points of the multiple reference measured points of the certain target measured point and the experimental variogram value of the certain target measured point and each reference measured point of the multiple reference measured points of the certain target measured point, a general Kriging equation set is established and solved to obtain multiple reference weight coefficients corresponding to the multiple reference measured points of the certain target measured point; According to the electrical field intensity measured value of the multiple reference measured points of the certain target measured point and the multiple reference weight coefficients, the electrical field intensity estimated value of the certain target measured point is calculated.

6. The power distribution network non-power-off operation risk early warning method according to claim 1, characterized in that, For a certain human joint of the multiple human joints, whether there is a human electric shock risk is judged according to the corresponding current electrical field intensity, including: For each of the plurality of human joints, a corresponding current electric field strength is determined , and a corresponding current voltage is calculated according to the following equation : In the formula, represents the insulation thickness of the insulation protector for electrically protecting the human body part corresponding to the certain human body joint. Determine the current voltage The system checks whether the insulation voltage of the insulating protective component is multiplied by a preset coefficient. If so, it determines that there is a risk of electric shock to the human body. The preset coefficient has a range of values. .

7. A device for early warning of risks in power distribution network operation without power interruption, characterized in that, It comprises a collection data acquisition module, an electrical field intensity association module, an electrical field distribution inversion module, a current field intensity determination module and a risk warning triggering module which are sequentially connected in communication; The collection data acquisition module is used for acquiring the Kinect skeletal data collected by a Kinect device in real time and the electrical field intensity collected by at least two MEMS electrical field sensors in real time, wherein the Kinect device is arranged at a distribution network non-power outage work site and makes the lens field of view cover the work area of the worker, the Kinect skeletal data contains the three-dimensional coordinates of the multiple human joints of the worker, the multiple human joints include any combination of the head joint, the neck joint, the spine shoulder joint, the middle spine joint, the base spine joint, the left shoulder joint, the left elbow joint, the left wrist joint, the left hand joint, the left hip joint, the left knee joint, the left ankle joint, the left foot joint, the right shoulder joint, the right elbow joint, the right wrist joint, the right hand joint, the right hip joint, the right knee joint, the right ankle joint and the right foot joint, the at least two MEMS electrical field sensors correspond to at least two human joints one by one, and the MEMS electrical field sensor is attached and arranged on the human body part corresponding to the joint; The electrical field intensity association module is used for, for each human joint of the at least two human joints, taking the electrical field intensity collected by the corresponding MEMS electrical field sensor at any historical time as the electrical field intensity measured value of the corresponding historical position at the any historical time. The electric field distribution inversion module is configured to obtain the three-dimensional distribution of the electric field at the distribution network live-line operation site based on the Kriging interpolation method according to the measured values of the electric field intensity at the multiple measured points, and specifically includes: calculating the semi-variance of the electric field intensity of each pair of measured points in the multiple measured points respectively to obtain the semi-variance values of the electric field intensity of multiple pairs of measured points according to the measured values of the electric field intensity at the multiple measured points, and further calculating the distance values of the multiple pairs of measured points according to the known coordinates of the multiple measured points, wherein the multiple measured points contain all historical positions of the at least two human body joints in the nearest historical period; determining a lag distance and the longest distance value in the distance values of the multiple pairs of measured points , wherein represents a positive number, represents a positive number greater than ; according to the lag distance , the first interval is divided into multiple first subintervals as follows: , wherein , represents a rounding-up function, represents a positive integer less than ; according to the relationship between the distance values of the multiple pairs of measured points and the multiple first subintervals, the multiple pairs of measured points are divided into multiple first groups corresponding to the multiple first subintervals one by one; according to the semi-variance values and distance values of the multiple pairs of measured points, the semi-variance average value and distance average value of each first group in the multiple first groups are calculated; according to the semi-variance average value and distance average value of each first group, the model coefficients of multiple experimental variogram models are fitted, wherein the semi-variance average value is used as an experimental variogram value in the fitting process, and the distance average value is used as the distance from the regionalized variable to the measured point in the fitting process; according to the measured values of the electric field intensity of the multiple measured points, error analysis is performed by applying the model coefficients of the multiple experimental variogram models to obtain the model quality evaluation index value of each experimental variogram model in the multiple experimental variogram models; according to the model quality evaluation index value of each experimental variogram model, the optimal experimental variogram model that best meets the model optimization preset condition is determined from the multiple experimental variogram models; according to the known coordinates of the multiple measured points and the known coordinates of a target measured point in the distribution network live-line operation site, the measured points located around the target measured point are determined from the multiple measured points, wherein represents a positive integer greater than 2; according to the known coordinates of the target measured point and the known coordinates of the measured points, the distance from the target measured point to the distance values of each of the measured points, and the distance values are taken as the distance of the regionalized variable to the target point into the optimal experimental variogram model, and then the model coefficients of the optimal experimental variogram model are applied to calculate the distance between the target point and the experimental variogram values of each of the measured points; according to the electrical field intensity semi-variance values of each pair of the measured points and the distance between the target point and the experimental variogram values of each of the measured points, the following ordinary Kriging equation group is established: In the formula, and They represent positive integers, Indicates the above The first measured point The weight coefficients corresponding to each measured point and to be solved. Indicates the first The measured points and the above The first measured point The semivariance of the electric field strength corresponding to each measured point Denotes the Lagrange multiplier factor to be solved. Indicates that the target measurement point and the first The experimental variogram values ​​at each measured point; the ordinary kriging equations are solved to obtain the values ​​corresponding to the experimental variogram values ​​at each measured point. Each measured point corresponds to one Each weighting coefficient; according to the The measured electric field strength values ​​at each measurement point are used to calculate the estimated electric field strength at the target measurement point using the following formula. : In the formula, represents the measured value of the electric field intensity of the first measured point; the estimated value of the electric field intensity of all the target measured points in the whole domain of the distribution network live working site is taken as the three-dimensional distribution of the electric field of the distribution network live working site. The current field intensity determination module is further communicatively connected to the collected data acquisition module, configured to determine a corresponding current electric field intensity for each of the plurality of human joint nodes according to the three-dimensional distribution of the electric field and the corresponding current location; The risk early warning triggering module is configured to determine whether there is a human body electric shock risk for each of the plurality of human joint nodes according to the corresponding current electric field intensity, and if so, trigger an early warning action.

8. A computer device, comprising: The device comprises a memory, a processor and a transceiver connected in sequence, wherein the memory is configured to store a computer program, the transceiver is configured to receive and send messages, and the processor is configured to read the computer program and execute the network distribution non-power operation risk early warning method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that The computer readable storage medium stores instructions, and when the instructions run on the computer, the network distribution non-power operation risk early warning method according to any one of claims 1-6 is executed.

Images