Safety situation awareness method for whole process of distribution network live working

Abstract

This application provides a method for full-process safety situation awareness in live-line work on power distribution networks, including: acquiring real-time three-dimensional coordinates of each joint of the worker from a multi-joint skeletal tracking unit deployed on an insulated bucket truck; extracting the real-time angles between the lines connecting the shoulder and elbow joints and the elbow and wrist joints as the elbow abduction angle; generating a set of three-dimensional coordinates of key contour points on the worker's body surface based on the real-time three-dimensional coordinates of each joint, combined with preset skin layer thickness and contour outward distance at each skeletal node; comparing the real-time elbow abduction angle with a preset first threshold, triggering a dynamic recognition process when the elbow abduction angle exceeds the first threshold; and fusing the safety distance monitoring data for the most prominent part with the conventional safety distance monitoring data for the head and shoulder area to obtain a comprehensive safety distance monitoring result.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for full-process safety situation awareness during live-line work in power distribution networks. Background Technology

[0002] Live-line work on power distribution networks is a high-risk task in power system maintenance. Its core objective is to complete close-range operations while ensuring the safety of personnel and equipment. Because the work environment involves high-voltage live conductors, even a slight loss of safe distance can lead to serious accidents. Therefore, real-time and accurate sensing of the safe distance between workers and live conductors is a crucial link in ensuring safety throughout the entire process. Currently, most safe distance monitoring methods primarily establish protection models based on the position of the worker's head and shoulders, assuming that as long as the head and shoulders maintain a sufficient distance from the live conductor, overall safety is guaranteed.

[0003] A current technology, CN115861407A, discloses a method and system for detecting safe distances based on deep learning. This method uses a deep learning model to identify live equipment and calculates the distance between the worker and the equipment based on a reference scale in an image. However, it suffers from a problem: it cannot capture in real-time changes in the worker's posture (such as elbow abduction) that cause the dynamic shift of the part closest to the live object, resulting in a monitoring blind spot. While this method works when the worker is in a relatively standard posture, in actual operation, the space inside the bucket of a boom truck is extremely limited, severely restricting the worker's range of motion. When workers need to retrieve or place insulating tools from tool bags or the inside of the bucket, or adjust their work position, the limited space forces them to extend their elbows significantly outwards, sometimes even completely. This abduction movement causes the protruding part of the body closest to the live object to rapidly shift from the head or shoulder to the tip of the elbow. Existing monitoring systems, which always use the head and shoulders as the closest point, cannot capture this instantaneous change in time, resulting in a monitoring blind spot in the elbow area. Especially in confined spaces, the intention is to reduce the risk of contact by restricting personnel movement. However, precisely because of the limited space, personnel are forced to extend their elbows further outward to perform necessary actions such as retrieving and placing tools, making the elbow the new most dangerous protruding part. The minimum safe distance no longer consistently occurs at the head and shoulder level, but dynamically shifts to the tip of the elbow as the elbow abduction angle increases instantaneously.

[0004] The mismatch between the changes in limb posture caused by spatial constraints and traditional fixed-benchmark monitoring makes it difficult for existing methods to achieve complete coverage of the entire human body in real-world work scenarios. Therefore, accurately identifying the moment when the elbow is significantly abducted due to actions such as picking up or placing tools in a confined space, and determining in real time the part of the human body that is actually closest to the live conductor at that moment, has become a key issue in solving the problem of insufficient safety distance monitoring coverage in the entire process of safety situation awareness for live-line work in power distribution networks. Summary of the Invention

[0005] This invention provides a method for full-process safety situation awareness during live-line working in distribution networks, including:

[0006] The real-time three-dimensional coordinates of each joint of the operator are obtained from the multi-joint skeletal tracking unit deployed on the insulated bucket truck. The real-time angle between the line connecting the shoulder joint and the elbow joint, and the line connecting the elbow joint and the wrist joint are extracted as the elbow abduction angle.

[0007] Based on the real-time three-dimensional coordinates of each joint of the worker, combined with the preset skin layer thickness and contour expansion distance at each bone node, a set of three-dimensional coordinates of key contour points on the worker's body surface is generated.

[0008] The original distance data of the surrounding charged bodies is collected by the omnidirectional laser ranging unit deployed on the insulated bucket truck. The original distance data of the surrounding charged bodies is fused with the three-dimensional coordinate set of key contour points on the body surface to extract the instantaneous distance set between each contour point of the human body and the surrounding charged bodies.

[0009] The real-time elbow abduction angle is compared with a preset first threshold. When the elbow abduction angle exceeds the first threshold, a dynamic recognition process is triggered.

[0010] After the dynamic recognition process is triggered, the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle are retrieved. The spatial boundary range of the area extending outward from the elbow joint and the subset of all contour points falling within the boundary are identified. The instantaneous distance value between each contour point in the contour point subset and the surrounding charged body is extracted. The contour point with the smallest instantaneous distance value is determined as the most prominent part of the human body at the current moment.

[0011] The instantaneous distances between the most prominent part of the human body and each charged object are extracted from the instantaneous distance set to generate safe distance monitoring data for the most prominent part. At the same time, the instantaneous distances between each contour point of the head and shoulder area and the charged object are extracted from the instantaneous distance set to generate regular safe distance monitoring data for the head and shoulder area.

[0012] The safety distance monitoring data for the most prominent parts is combined with the routine safety distance monitoring data for the head and shoulder areas to obtain a comprehensive safety distance monitoring result.

[0013] Furthermore, the step of acquiring the real-time three-dimensional coordinates of each joint of the operator collected by the multi-joint skeletal tracking unit deployed on the insulated bucket truck, and extracting the real-time angles between the lines connecting the shoulder and elbow joints and the lines connecting the elbow and wrist joints as the elbow abduction angle, includes:

[0014] Obtain the real-time three-dimensional coordinates of the shoulder, elbow, and wrist joints of the operator;

[0015] Generate the upper arm bone connection vector and the forearm bone connection vector;

[0016] The vector connecting the upper arm bones and the vector connecting the forearm bones are processed to obtain the real-time angle between the two vectors, and this angle is determined as the elbow abduction angle at the current moment.

[0017] Furthermore, based on the real-time three-dimensional coordinates of each joint of the worker, combined with the preset skin layer thickness and contour expansion distance at each skeletal node, a set of three-dimensional coordinates of key contour points on the worker's body surface is generated, including:

[0018] Obtain the real-time three-dimensional coordinates of each joint of the operator;

[0019] Retrieve the skin layer thickness and contour outward distance values ​​corresponding to each skeletal node from the preset human body surface parameter library;

[0020] For each skeletal node, the radial direction pointing outward from the human body is determined based on the vertical direction of the line connecting adjacent bones. The skin layer thickness value and the contour outward expansion distance value are superimposed to obtain the comprehensive outward expansion amount.

[0021] Using the three-dimensional coordinates of the skeletal node as a reference point, the coordinates are offset along the radial direction according to the comprehensive outward expansion to obtain the three-dimensional coordinates of the body surface contour point corresponding to the node.

[0022] Traverse all skeletal nodes to generate the 3D coordinates of the body surface contour points corresponding to each node, forming a set of 3D coordinates of the key contour points on the worker's body surface.

[0023] Furthermore, the omnidirectional laser ranging unit deployed on the insulated bucket truck collects raw distance data from surrounding charged bodies, fuses this raw distance data with the three-dimensional coordinate set of key contour points on the body surface, and extracts the instantaneous distance set between each contour point of the human body and surrounding charged bodies, including:

[0024] The omnidirectional laser ranging unit collects raw distance data and corresponding azimuth angles in all directions within the surrounding space.

[0025] The reflection points on the surfaces of each charged body are converted into three-dimensional rectangular coordinates, thus obtaining the three-dimensional coordinates of the reflection points on the surfaces of the surrounding charged bodies.

[0026] Spatial registration is performed by placing the three-dimensional coordinates of the key contour points on the body surface and the three-dimensional coordinates of the reflection points on the surface of each charged body in the same coordinate system.

[0027] For each key contour point on the body surface, calculate the distance between the contour point and the reflection point on the surface of each charged body to obtain the instantaneous distance value between the contour point and the surrounding charged bodies.

[0028] Traverse all contour points to form a set of instantaneous distances between each contour point of the human body and the surrounding charged bodies.

[0029] Furthermore, the real-time elbow abduction angle is compared with a preset first threshold. When the elbow abduction angle exceeds the first threshold, a dynamic recognition process is triggered, including:

[0030] Obtain the real-time elbow abduction angle at the current moment;

[0031] Read the first threshold and compare the real-time elbow abduction angle with the first threshold.

[0032] If the real-time elbow abduction angle exceeds the first threshold, it is determined that the current worker's elbow is in a state of large abduction, triggering the dynamic recognition process.

[0033] Furthermore, after the dynamic recognition process is triggered, the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle are retrieved. The spatial boundary range of the area extending outward from the elbow joint and the subset of all contour points falling within this boundary are identified. The instantaneous distance values ​​between each contour point in the subset and surrounding charged bodies are extracted, and the contour point with the smallest instantaneous distance value is determined as the most prominent part of the human body at the current moment, including:

[0034] Retrieve the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle;

[0035] Obtain the three-dimensional coordinates of the elbow joint as the starting point of the spatial boundary;

[0036] Based on the angle between the line connecting the upper arm bones and the line connecting the forearm bones, the fan-shaped area extending outward from the elbow joint is determined as the spatial boundary.

[0037] Traverse the three-dimensional coordinate set of the key contour points on the body surface, determine whether each contour point falls within the spatial boundary range, and include the contour points that fall within the boundary range into the contour point subset.

[0038] Extract the instantaneous distance values ​​between each contour point in the subset of contour points and the surrounding charged bodies from the instantaneous distance set;

[0039] The instantaneous distance values ​​of all contour points within the subset of contour points are numerically sorted.

[0040] Extract the contour point with the smallest instantaneous distance value and determine that contour point as the most prominent part of the human body at the current moment.

[0041] Furthermore, the process of retrieving the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle, and identifying the spatial boundary range of the area extending outward from the elbow joint and the subset of contour points falling within that boundary, includes:

[0042] Retrieve the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle;

[0043] Obtain the three-dimensional coordinates of the elbow joint;

[0044] Based on the elbow abduction angle and the direction of the line connecting the upper arm bones, determine the spatial boundary range extending outward from the elbow joint to the outside of the human body.

[0045] Traverse the three-dimensional coordinate set of the key contour points on the body surface, filter out all contour points that fall within the spatial boundary range, and form a subset of contour points.

[0046] Furthermore, the instantaneous distances between the most prominent part of the human body and each charged object are extracted from the instantaneous distance set to generate safe distance monitoring data for the most prominent part. Simultaneously, the instantaneous distances between each contour point in the head and shoulder region and the charged object are extracted from the instantaneous distance set to generate conventional safe distance monitoring data for the head and shoulder region, including:

[0047] Obtain the outline point number of the most prominent part of the human body at the current moment;

[0048] Extract the instantaneous distance values ​​between the contour points of the most prominent part and each surrounding charged body, as well as the corresponding charged body position information, from the instantaneous distance set. Then, associate and combine the instantaneous distance values ​​and the charged body position information to generate safe distance monitoring data for the most prominent part.

[0049] Based on the preset range of outline point numbers for the head and shoulder region, determine all outline points belonging to the head and shoulder region from the three-dimensional coordinate set of key outline points on the body surface.

[0050] Extract the instantaneous distance values ​​between each contour point in the head and shoulder region and the surrounding charged bodies from the instantaneous distance set to generate routine safe distance monitoring data for the head and shoulder region;

[0051] The safe distance monitoring data for the most prominent part is output in parallel with the regular safe distance monitoring data for the head and shoulder area.

[0052] Furthermore, the step of obtaining the contour point number of the most prominent part of the human body at the current moment, and extracting the instantaneous distance values ​​between the contour points of the most prominent part and each surrounding charged body, as well as the corresponding charged body position information, from the instantaneous distance set, includes:

[0053] Obtain the three-dimensional coordinates of the most prominent part of the human body at the current moment and the corresponding contour point number;

[0054] Based on the contour point number, extract the instantaneous distance values ​​between the point and each surrounding charged body from the instantaneous distance set;

[0055] Obtain the spatial coordinates of each charged body;

[0056] The three-dimensional coordinates and instantaneous distance values ​​of the most prominent part are correlated with the spatial coordinates of each charged body to form safe distance monitoring data that includes the coordinates of the most prominent part, the position of the nearest charged body, and the instantaneous distance value.

[0057] Furthermore, the safety distance monitoring data for the most prominent part is fused with the routine safety distance monitoring data for the head and shoulder area to obtain a comprehensive safety distance monitoring result, including:

[0058] Acquire safe distance monitoring data for the most prominent areas and routine safe distance monitoring data for the head and shoulder areas;

[0059] The monitoring data is integrated and merged according to human body parts to form a monitoring data set with a unified format;

[0060] Based on the instantaneous distance values ​​of each part in the monitoring data set, the safe distance monitoring status covering the most prominent part and the head and shoulder area is determined, and a comprehensive safe distance monitoring result is obtained.

[0061] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0062] This invention discloses a method for full-process safety situation awareness in live-line work on power distribution networks. By acquiring the three-dimensional coordinates of the worker's joints and calculating the real-time elbow abduction angle, a dynamic identification process is triggered when the angle exceeds a threshold. This invention integrates preset skin thickness and contour expansion data to generate a set of three-dimensional coordinates of key contour points on the body surface, and combines the distance data of the live body collected by the omnidirectional laser ranging unit to extract the instantaneous distance set between the human contour points and the live body. In the dynamic identification process, this invention identifies the outer spatial boundary and contour point subset starting from the elbow joint based on the current elbow abduction angle. By comparing the distance between each point in the subset and the live body, the most prominent part of the human body at the current moment is accurately determined, and then a fusion comprehensive safety distance monitoring result is generated for the most prominent part and the head and shoulder area. This achieves proactive and accurate perception and early warning of the risk point closest to the live body under the dynamic dangerous posture of the worker. Attached Figure Description

[0063] Figure 1 This is a flowchart of the full-process safety situation awareness method for live-line working in distribution networks according to the present invention.

[0064] Figure 2 This is a schematic diagram of the full-process safety situation awareness method for live-line working in distribution networks according to the present invention.

[0065] Figure 3 This is another schematic diagram of the full-process safety situation awareness method for live-line work in distribution networks according to the present invention. Detailed Implementation

[0066] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0067] like Figures 1-3 The safety situation awareness method for the entire process of live-line working in the distribution network in this embodiment may specifically include:

[0068] Step S101: Obtain the real-time three-dimensional coordinates of each joint of the operator from the multi-joint skeletal tracking unit deployed on the insulated bucket truck, and extract the real-time angle between the line connecting the shoulder joint and the elbow joint, and the line connecting the elbow joint and the wrist joint as the elbow abduction angle.

[0069] A multi-joint skeletal tracking unit deployed on an insulated bucket truck acquires the real-time three-dimensional coordinates of the operator's shoulder, elbow, and wrist joints. Based on the three-dimensional coordinates of the shoulder and elbow joints, an upper arm skeletal connection vector is generated; based on the three-dimensional coordinates of the elbow and wrist joints, a forearm skeletal connection vector is generated. The upper arm and forearm skeletal connection vectors are processed using a vector dot product operation to obtain the ratio of the dot product value to the product of the magnitudes of the two vectors. An inverse cosine operation is then performed on this ratio to obtain the real-time angle between the two vectors. This angle value is determined as the elbow abduction angle at the current moment.

[0070] In live-line work scenarios, the bucket of the insulated bucket truck is equipped with a multi-joint skeletal tracking unit. This tracking unit uses a depth vision sensor to capture the limb bones of the worker in real time.

[0071] For example, the depth vision sensor acquires the spatial position information of each joint of the worker's body by emitting structured light or by time-of-flight ranging, and outputs the real-time position data of the shoulder joint, elbow joint and wrist joint in the form of three-dimensional coordinates.

[0072] In one possible implementation, the process of generating the upper arm bone connection vector is as follows: taking the three-dimensional coordinates of the shoulder joint as the starting point and the three-dimensional coordinates of the elbow joint as the ending point, calculating the spatial displacement components between the two points to form a three-dimensional vector representing the direction of the upper arm; the process of generating the forearm bone connection vector is similar, constructing a three-dimensional vector with the three-dimensional coordinates of the elbow joint as the starting point and the three-dimensional coordinates of the wrist joint as the ending point.

[0073] Specifically, the vector dot product operation is performed by multiplying the corresponding coordinate components of the vector connecting the upper arm bones and the vector connecting the forearm bones, and then summing the results to obtain the dot product value. Simultaneously, the magnitudes of the two vectors are calculated separately and multiplied to obtain the magnitude product. The dot product value is then divided by the magnitude product to obtain the cosine value. This cosine value is then converted to an angle by performing an inverse cosine operation, yielding the elbow abduction angle. This angle reflects the degree of bending between the upper arm and forearm at the worker's elbow joint. As the worker retrieves or places tools in the insulated bucket, the elbow abduction angle changes in real time with the limb movements.

[0074] Step S102: Based on the real-time three-dimensional coordinates of each joint of the operator, and combined with the preset skin layer thickness and contour expansion distance at each bone node, generate a set of three-dimensional coordinates of key contour points on the operator's body surface.

[0075] The system acquires the real-time 3D coordinates of each joint of the worker, determines the spatial position of each skeletal node based on the human skeletal topology, and retrieves the corresponding skin thickness and contour outward distance values ​​from a pre-set human body surface parameter library. The skin thickness value represents the radial distance from the center of the bone to the outer layer of the skin, and the contour outward distance value represents the additional thickness from the outer layer of the skin to the outer edge of clothing or protective equipment. For each skeletal node, the radial direction pointing outward from the node is determined based on the perpendicular direction of the line connecting adjacent bones. The skin thickness value and the contour outward distance value are superimposed to obtain the comprehensive outward expansion of the node. Using the 3D coordinates of the skeletal node as a reference point, the coordinates are offset radially according to the comprehensive outward expansion to obtain the 3D coordinates of the corresponding body surface contour point. All skeletal nodes are traversed, and the 3D coordinates of the corresponding body surface contour points are generated one by one. The 3D coordinates of all body surface contour points are then aggregated to form a set of 3D coordinates of key contour points on the worker's body surface.

[0076] In live-line work scenarios, workers wearing insulated protective gear enter the bucket of an insulated bucket truck to perform operations. There is a certain thickness difference between the actual outer contour of different body parts and the skeletal positions. The human body surface parameter database pre-stores the skin layer thickness values ​​and contour outward distance values ​​at each skeletal node. The skin layer thickness values ​​are determined based on human anatomical measurements, while the contour outward distance values ​​are determined based on the thickness of the insulated gloves, insulated clothing, and other protective gear worn by the workers.

[0077] For example, the skin thickness value at the elbow joint represents the radial distance from the center of the elbow joint bone to the skin surface, while the outward expansion distance value represents the additional thickness from the skin surface to the outer edge of the cuff of the insulating garment.

[0078] In one possible implementation, for the elbow joint node, based on the plane formed by the line connecting the upper arm bones and the line connecting the forearm bones, the direction perpendicular to the plane and pointing outward of the body is taken as the radial direction. The skin layer thickness value and the contour expansion distance value are added along the radial direction to obtain the comprehensive expansion amount. The distance of the comprehensive expansion amount is offset along the radial direction from the three-dimensional coordinates of the elbow joint to obtain the three-dimensional coordinates of the body surface contour point corresponding to the elbow joint.

[0079] Specifically, the human skeletal topology includes multiple nodes such as the shoulder joint, elbow joint, wrist joint, hip joint, and knee joint. Each node determines its radial direction and performs coordinate offset in the manner described above.

[0080] It should be noted that the overall outward expansion varies at different skeletal nodes. At the elbow joint tip, due to the prominent bone and thin subcutaneous fat, the skin layer thickness is relatively small, while at the shoulder joint, due to the thicker muscle tissue, the skin layer thickness is relatively large. After completing the coordinate offset of all skeletal nodes, the three-dimensional coordinates of the corresponding body surface contour points are aggregated to form a three-dimensional coordinate set. Each contour point in this set can represent the spatial distribution of the worker's body surface outer edge in the current posture.

[0081] Step S103: The original distance data of the surrounding charged bodies is collected by the omnidirectional laser ranging unit deployed on the insulated bucket truck. The original distance data of the surrounding charged bodies is fused with the three-dimensional coordinate set of key contour points on the body surface to extract the instantaneous distance set between each contour point of the human body and the surrounding charged bodies.

[0082] An omnidirectional laser ranging unit deployed on an insulated bucket truck periodically rotates and scans along the horizontal and vertical directions, using the center of the bucket as the scanning reference point. Reflected signals from various directions in the surrounding space are collected. Based on the time difference between laser emission and signal reception, the original distance data from the scanning reference point to the surface of surrounding charged bodies and the corresponding azimuth angles are obtained in each scanning direction. The fixed installation position coordinates of the scanning reference point of the omnidirectional laser ranging unit in the insulated bucket truck coordinate system are obtained. Based on the original distance data and corresponding azimuth angles, the reflection points on the surfaces of each charged body are converted from polar coordinates to three-dimensional rectangular coordinates, obtaining the three-dimensional coordinates of the reflection points on the surrounding charged body surfaces. The three-dimensional coordinate set of key contour points on the body surface is retrieved, and the three-dimensional coordinates of the key contour points and the three-dimensional coordinates of the reflection points on the surfaces of each charged body are spatially registered in the same insulated bucket truck coordinate system. For each key contour point, the Euclidean distance between the contour point and the reflection points on the surfaces of each charged body is calculated, obtaining the instantaneous distance value between the contour point and the surrounding charged bodies. Traverse all contour points in the three-dimensional coordinate set of key contour points on the body surface, calculate the instantaneous distance between each contour point and the surrounding charged body, and aggregate all instantaneous distance values ​​to form a set of instantaneous distances between each contour point of the human body and the surrounding charged body.

[0083] In live-line working scenarios, multiple live conductors and live equipment are distributed around the bucket of an insulated bucket truck. When workers operate inside the bucket, the spatial distance between their body parts and the surrounding live conductors is dynamically changing. An omnidirectional laser ranging unit is installed on the top or side wall of the bucket of the insulated bucket truck. Using a fixed installation position as the scanning reference point, a mechanical rotation mechanism drives the laser emitter to rotate continuously in both horizontal and vertical directions, achieving omnidirectional coverage scanning of the space surrounding the bucket.

[0084] For example, the omnidirectional laser ranging unit rotates gradually in the horizontal direction at preset angular intervals, and then performs a vertical elevation scan at each horizontal angular position. The laser beam is directed into the surrounding space and is reflected when it encounters the surfaces of objects such as live wires, insulators, and crossarms. The basic principle of laser ranging is to calculate the distance based on the flight time of the laser pulse from emission to reception of the reflected signal. The speed of laser propagation in air is a known constant, and half of the flight time multiplied by the speed of light gives the radial distance from the scanning reference point to the reflection point.

[0085] In one possible implementation, after completing a full periodic rotational scan, the omnidirectional laser ranging unit outputs point cloud data consisting of several measurement points. Each measurement point includes its radial distance relative to the scanning reference point, as well as its corresponding horizontal azimuth and vertical pitch angles. The raw distance data is the radial distance value of each measurement point, and the azimuth angle includes two angular components: horizontal azimuth and vertical pitch angle.

[0086] Specifically, when converting measurement data in polar coordinates to three-dimensional rectangular coordinates, a local coordinate system is established with the scanning reference point as the origin. Based on the radial distance, horizontal azimuth angle, and vertical pitch angle, the three-dimensional coordinates of each charged body surface reflection point in the local coordinate system are calculated through the geometric transformation relationship from spherical coordinates to rectangular coordinates. Furthermore, based on the fixed installation position coordinates of the omnidirectional laser ranging unit's scanning reference point in the insulated bucket truck coordinate system, the three-dimensional coordinates of each reflection point are translated and transformed to uniformly represent them in the insulated bucket truck coordinate system.

[0087] It should be noted that the three-dimensional coordinate set of key contour points on the body surface originates from the real-time capture of the worker's limb posture by the skeletal tracking unit. This coordinate set also uses the insulated bucket truck coordinate system as a reference. The spatial registration process involves aligning two sets of three-dimensional coordinate data from different sources within the same coordinate system. Since both the omnidirectional laser ranging unit and the skeletal tracking unit are mounted on the same insulated bucket truck with fixed installation positions, there is a definite rigid transformation relationship between their coordinate systems. Registration can be completed using a pre-calibrated transformation matrix.

[0088] In one embodiment, for any one of the three-dimensional coordinates of key contour points on the body surface, the three-dimensional coordinates of all reflection points on the surface of the charged body are traversed, and the Euclidean distance between the contour point and each reflection point is calculated. The Euclidean distance is calculated as the square root of the sum of the squares of the differences between the coordinate components of the two points in three-dimensional space, and this distance value represents the straight-line spatial distance between the contour point and the surface of the charged body.

[0089] It is understandable that, since charged conductors and charged equipment are distributed linearly or planarly in space, the reflection points obtained by the omnidirectional laser ranging unit are actually discrete sampling points on the surface of the charged body. The instantaneous distance between a contour point and the surrounding charged bodies is taken as the minimum value of the Euclidean distance between the contour point and all reflection points on the surface of the charged bodies. This minimum value represents the actual spatial distance of the contour point from the nearest surface of the charged body.

[0090] Preferably, the instantaneous distance set is first defined as a set formed by traversing all key contour points on the body surface and calculating the instantaneous distance value between each contour point and the surrounding charged body, and then arranging these instantaneous distance values ​​in an orderly manner according to the contour point number. Each element in this set corresponds to the shortest distance between a body surface contour point and the surrounding charged body, and the set as a whole reflects the spatial distance distribution between various parts of the worker's body and the surrounding charged environment.

[0091] Step S104: Compare the real-time elbow abduction angle with a preset first threshold. When the elbow abduction angle exceeds the first threshold, trigger the dynamic recognition process.

[0092] The system acquires the real-time elbow abduction angle and reads a first threshold from a preset parameter storage area. This first threshold represents a critical judgment value for the elbow abduction angle. The real-time elbow abduction angle is then compared with the first threshold to obtain a comparison result. If the comparison result shows that the real-time elbow abduction angle exceeds the first threshold, it is determined that the current worker's elbow is in a state of significant abduction, triggering a dynamic recognition process.

[0093] During live-line work on power distribution networks, when workers retrieve or place tools or adjust their work position inside the insulated bucket, their elbows will abduct to varying degrees. The first threshold is a critical value for the elbow abduction angle pre-set according to ergonomics and work safety regulations. When the elbow abduction angle is below this critical value, the most prominent part of the body is usually still the head or shoulder; when the elbow abduction angle exceeds this critical value, the tip of the elbow begins to protrude outward and gradually becomes the part of the body closest to the live conductor.

[0094] For example, the value of the first threshold is calibrated based on the space size inside the insulating bucket and the distribution of the surrounding charged bodies. When the worker is in a normal standing posture, the elbow is naturally drooping with a small abduction angle, so there is no need to trigger the dynamic recognition process. When the worker reaches for or puts down a tool, causing the elbow to abduct significantly, the real-time elbow abduction angle exceeds the first threshold. It is determined that the elbow is currently in a state of significant abduction, and the dynamic recognition process is triggered to focus on monitoring the elbow area.

[0095] It should be noted that the comparison process is executed periodically at preset time intervals, and each update of the real-time elbow abduction angle is compared with the first threshold, thereby achieving continuous tracking of the changes in the elbow posture of the operator.

[0096] Step S105: After the dynamic recognition process is triggered, retrieve the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle, identify the spatial boundary range of the area extending outward from the elbow joint as the starting point and the subset of all contour points falling within the boundary, extract the instantaneous distance value between each contour point in the contour point subset and the surrounding charged body, and determine the contour point with the smallest instantaneous distance value as the most prominent part of the human body at the current moment.

[0097] After the dynamic recognition process is triggered, the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle are retrieved. The three-dimensional coordinates of the elbow joint are obtained as the starting point of the spatial boundary. Based on the angle between the line connecting the upper arm bones and the line connecting the forearm bones, a fan-shaped area extending outward from the elbow joint is determined as the spatial boundary range. The three-dimensional coordinate set of the key contour points on the body surface is traversed. For each contour point, it is determined whether its three-dimensional coordinates fall within the spatial boundary range. If they do, the contour point is included in the contour point subset, resulting in a contour point subset containing all contour points in the elbow abduction area. The instantaneous distance values ​​between each contour point in the contour point subset and surrounding charged bodies are extracted from the instantaneous distance set. The instantaneous distance values ​​of all contour points in the contour point subset are numerically sorted to obtain an ascending sequence of instantaneous distance values. The contour point with the smallest instantaneous distance value is extracted from the sequence and identified as the most prominent part of the body at the current moment.

[0098] In live-line working scenarios on power distribution networks, when the elbow abduction angle of the worker exceeds a first threshold, the dynamic identification process is triggered. At this time, the elbow area becomes the key area of ​​focus for safe distance monitoring. The core task of the dynamic identification process is to identify the contour points located in the elbow abduction area from the contour points on the entire body surface, and determine the contour point closest to the live conductor as the most prominent part of the human body at that moment.

[0099] For example, the three-dimensional coordinate set of key contour points on the body surface includes the coordinates of contour points corresponding to each skeletal node of the worker's body, wherein the contour point corresponding to the elbow joint is located at the intersection of the upper arm and forearm. When the elbow abducts, the contour points at the elbow joint and the contour points on the outer side of the forearm protrude outwards from the body, and these contour points constitute the spatial range of the elbow abduction area.

[0100] In one possible implementation, the determination of the spatial boundary range is as follows: Using the three-dimensional coordinates of the elbow joint as the vertex, the direction of the line connecting the upper arm bones and the direction of the line connecting the forearm bones jointly define a plane. On this plane, with the elbow joint as the vertex and the angle between the lines connecting the upper arm bones and the forearm bones as the subdivision angle, a fan-shaped region extends outward from the body. The size of the subdivision angle of this fan-shaped region is related to the current elbow abduction angle; the larger the abduction angle, the wider the coverage area of ​​the fan-shaped region. Further, this fan-shaped region is stretched along a direction perpendicular to the aforementioned plane to form a three-dimensional cone-shaped region, which is the spatial boundary range extending outward from the elbow joint.

[0101] Specifically, the method for determining whether a contour point falls within the spatial boundary range is as follows: Calculate the spatial position vector of the contour point relative to the elbow joint, project this position vector onto the plane defined by the lines connecting the upper arm bones and the forearm bones, and determine whether the projected vector is within the angular range of the fan-shaped region. Simultaneously, determine whether the vertical distance between the contour point and the aforementioned plane is less than a preset thickness threshold. If both conditions are met, the contour point is determined to fall within the spatial boundary range and is included in the contour point subset.

[0102] It should be noted that the contour points included in the contour point subset are mainly distributed at the tip of the elbow joint, the outer side of the forearm, and the outer side of the upper arm near the elbow joint. These contour points protrude outwards when the elbow is abducted, representing potentially dangerous areas close to nearby charged objects. By constructing a contour point subset, the scope of safe distance monitoring is narrowed from the entire body to the elbow abduction area, enabling precise monitoring of key areas.

[0103] In one embodiment, the instantaneous distances between the worker's entire body contour points and surrounding charged objects are calculated through previous steps and stored as an instantaneous distance set. When extracting the instantaneous distance values ​​between each contour point within a subset of contour points and charged objects from this instantaneous distance set, an index lookup is performed in the instantaneous distance set based on the contour point's number to obtain the instantaneous distance value corresponding to each contour point. Since the instantaneous distance set records the distances between all contour points on the entire body and charged objects, the instantaneous distance values ​​corresponding to the contour point subset, as a subset of the entire body contour points, can also be directly extracted from the instantaneous distance set.

[0104] Preferably, when numerically sorting the instantaneous distance values ​​of all contour points within the contour point subset, an ascending order is used, so that the contour point with the smallest instantaneous distance value is placed at the beginning of the sequence. After sorting, the contour point at the beginning of the sequence is the contour point in the contour point subset that is closest to the charged body, and this contour point is determined as the most prominent part of the human body at the current moment.

[0105] Understandably, when a worker extends their elbow significantly to retrieve or place tools within the insulated container, the contour point at the tip of the elbow joint typically becomes the contour point with the smallest instantaneous distance value. At this moment, the most prominent part of the human body shifts from the head and shoulders to the tip of the elbow. By executing the dynamic recognition process in real time, this instantaneous change can be accurately captured, avoiding monitoring blind spots caused by fixed monitoring of the head and shoulders.

[0106] Retrieve the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle, identify the spatial boundary range of the area extending outward from the elbow joint and the subset of all contour points falling within the boundary, extract the instantaneous distance values ​​between each contour point in the subset of contour points and the surrounding charged bodies, and determine the contour point with the smallest instantaneous distance value as the most prominent part of the human body at the current moment.

[0107] The three-dimensional coordinates of key contour points on the body surface are retrieved along with the current elbow abduction angle to obtain the three-dimensional coordinates of the elbow joint. Based on the direction of the line connecting the elbow abduction angle and the upper arm bones, the spatial boundary range extending outward from the elbow joint is determined. The three-dimensional coordinates of the key contour points on the body surface are traversed, and all contour points falling within the spatial boundary range are selected to form a subset of contour points. The instantaneous distance values ​​between each contour point in the subset and surrounding charged bodies are extracted from the instantaneous distance set. The instantaneous distance values ​​of all contour points in the subset are compared to obtain the contour point with the smallest instantaneous distance value. The contour point with the smallest instantaneous distance value is determined as the most prominent part of the human body at the current moment.

[0108] In live-line working scenarios, when the worker's elbow is in a significantly abducted position, the outline of the elbow area becomes a potentially dangerous location close to live conductors. The spatial boundary is determined based on the spatial position of the elbow joint and the elbow abduction angle. Centered on the three-dimensional coordinates of the elbow joint, the boundary extends outward along the angle determined by the line connecting the upper arm bones and the line connecting the forearm bones, forming a three-dimensional spatial range covering the elbow abduction area.

[0109] For example, the shape of the spatial boundary can be a cone-shaped region with the elbow joint as its apex, the axis of which extends along the elbow abduction direction, and the angle of the cone-shaped region is associated with the current elbow abduction angle. When the elbow abduction angle is large, the coverage area of ​​the cone-shaped region expands accordingly to include more contour points located on the outer side of the elbow.

[0110] In one possible implementation, when traversing the three-dimensional coordinate set of key contour points on the body surface, the spatial positional relationship of each contour point relative to the elbow joint is calculated, and it is determined whether the contour point is within the spatial boundary range. If the contour point is within the boundary range, it is added to the contour point subset; if the contour point is outside the boundary range, it is skipped and the traversal continues to the next one. After the traversal is completed, the contour point subset contains all contour points in the elbow tip, the outer forearm, and the outer upper arm near the elbow joint. First, the instantaneous distance values ​​of each contour point in the contour point subset are extracted from the instantaneous distance set calculated in the previous steps. This set records the real-time distance between all contour points and the target. The corresponding distance record is searched in the instantaneous distance set according to the contour point's index. All extracted instantaneous distance values ​​are numerically compared, and the instantaneous distance value with the smallest value and its corresponding contour point are found.

[0111] It should be noted that the contour point with the smallest instantaneous distance value represents the position closest to the surrounding charged body in the contour point subset, and this contour point is determined as the most prominent part of the human body at the current moment. Specifically, based on the three-dimensional coordinate set C of key contour points on the body surface and the current elbow abduction angle θ, the boundary of the outer extension region starting from the elbow joint point E is determined. Taking the elbow joint E as the vertex, the axial direction of the cone region is determined according to the direction of the angle bisector of the vector connecting the upper arm bones and the vector connecting the forearm bones. The half-angle α of the cone region is set to half of the current elbow abduction angle θ, i.e., α = θ / 2. For any contour point P in the three-dimensional coordinate set C, the angle β between the vector EP and the cone axis is calculated. If β ≤ α, the contour point P is determined to fall within the spatial boundary range and is included in the contour point subset S. From the contour point subset S, the Euclidean distance between each point P and the nearest reflection point B on the surface of the charged body is calculated. The contour point with the smallest d is selected as the most prominent part of the human body at the current moment. The spatial position of this most prominent part is dynamically updated according to the real-time changes in the elbow posture of the operator. When the elbow is retracted or the abduction angle decreases, the most prominent part may shift from the elbow area to the shoulder area.

[0112] Step S106: Extract the instantaneous distance between the most prominent part of the human body and each charged object from the instantaneous distance set to generate safe distance monitoring data for the most prominent part. At the same time, extract the instantaneous distance between each contour point of the head and shoulder area and the charged object from the instantaneous distance set to generate regular safe distance monitoring data for the head and shoulder area.

[0113] The system obtains the outline point number of the most prominent part of the human body at the current moment. It then extracts the instantaneous distance values ​​between the outline point of the most prominent part and each surrounding charged object, along with the corresponding position information of the charged objects, from the instantaneous distance set. These instantaneous distance values ​​and the position information of the charged objects are then correlated and combined to generate safe distance monitoring data for the most prominent part. Based on a preset range of outline point numbers for the head and shoulder region, all outline points belonging to the head and shoulder region are determined from the three-dimensional coordinate set of key outline points on the body surface. The instantaneous distance values ​​between each outline point in the head and shoulder region and surrounding charged objects are then extracted from the instantaneous distance set to generate regular safe distance monitoring data for the head and shoulder region. Finally, the safe distance monitoring data for the most prominent part and the regular safe distance monitoring data for the head and shoulder region are output in parallel.

[0114] In live-line working scenarios on power distribution networks, safety distance monitoring needs to simultaneously focus on the most prominent and dynamically changing parts, as well as the head and shoulder area monitored conventionally. The safety distance monitoring data for the most prominent parts includes the three-dimensional coordinates of the outline point of the most prominent part, the instantaneous distance values ​​between the outline point and each surrounding live object, and the spatial position information of each live object. The correlation and combination of these information constitutes a complete monitoring record.

[0115] For example, when the worker's elbow is abducted outwards, the most prominent part is identified as the outline point at the tip of the elbow joint. An instantaneous distance set is generated by calculating the distance between the outline point of the human body and each live wire around the insulating bucket in real time. The distance value between the outline point and each live wire is extracted from the set, and the spatial position coordinates of each live wire are recorded to form a complete monitoring record containing the outline point coordinates, distance values ​​and the position of the live body.

[0116] In one possible implementation, the range of outline point numbers for the head and shoulder region is a fixed numbering interval pre-determined based on the topology of the human skeleton, including all outline point numbers for the top of the head, forehead, outer side of the shoulder joint, and outer side of the upper arm near the shoulder joint. Outline points for the head and shoulder region are selected from the three-dimensional coordinate set of key outline points on the body surface based on this numbering range, and the instantaneous distance values ​​between each outline point and surrounding charged bodies are extracted from a pre-generated set of instantaneous distances based on the numbers of these outline points.

[0117] Specifically, routine safety distance monitoring data for the head and shoulder area reflects the spatial distance between the worker's head and shoulders and surrounding live conductors. When the worker's elbows are not significantly abducted, the head and shoulder area is typically the closest part of the human body to live conductors, making routine safety distance monitoring data the primary basis for determining safe distances.

[0118] It should be noted that the safety distance monitoring data for the most prominent body part and the routine safety distance monitoring data for the head and shoulder area are output in parallel. The two types of monitoring data are independent of each other and are valid simultaneously. When the dynamic recognition process is triggered, the monitoring data for the most prominent body part provides key monitoring information for the elbow abduction area, while the monitoring data for the head and shoulder area continues to provide routine monitoring information. Together, they constitute a set of safety distance monitoring data covering key parts of the human body.

[0119] Obtain the three-dimensional coordinates and corresponding contour point number of the most prominent part of the human body at the current moment, and extract the distance value between the contour point and each surrounding charged body and the position of the charged body from the instantaneous distance set. Establish a real-time safe distance monitoring record for the most prominent part, and form safe distance monitoring data containing the coordinates of the most prominent part, the position of the nearest charged body and the instantaneous distance value.

[0120] The system obtains the three-dimensional coordinates and corresponding contour point number of the most prominent part of the human body at the current moment. Based on this contour point number, it extracts the instantaneous distance values ​​between this point and each surrounding charged object from the aforementioned instantaneous distance set, and simultaneously obtains the spatial position coordinates of each charged object, thus obtaining the distance and position correlation data between the most prominent part and each charged object. Finally, it integrates the three-dimensional coordinates of the most prominent part, the extracted instantaneous distance values, and the spatial position coordinates of each charged object to form safe distance monitoring data that includes the coordinates of the most prominent part, the position of the nearest charged object, and the instantaneous distance value.

[0121] In live-line working scenarios on power distribution networks, the safety distance monitoring data is a complete record of the spatial distance relationship between the most prominent part of the human body and surrounding live conductors. After the dynamic identification process determines the most prominent part of the human body at the current moment, the corresponding distance record is extracted from the instantaneous distance set according to the number of the contour point, and the instantaneous distance value between the contour point and each live conductor, insulator and crossarm around the insulator bucket is obtained.

[0122] For example, the safety distance monitoring data includes three core information fields: the three-dimensional coordinates of the most prominent part are used to identify the spatial position of the human body closest to the charged body, the instantaneous distance value is used to characterize the actual spatial distance between the position and the charged body, and the spatial position coordinates of the charged body are used to determine the location of the nearest charged body.

[0123] In one possible implementation, when multiple charged bodies exist around the most prominent part, the smallest distance value and its corresponding charged body location are selected from all instantaneous distance values. This smallest distance value and the corresponding charged body location are then integrated with the coordinates of the most prominent part to form a complete safety distance monitoring data record. This monitoring data record can intuitively reflect the spatial distance between the most prominent part of the human body and the nearest charged body at the current moment.

[0124] Step S107: The safety distance monitoring data for the most prominent part is fused with the regular safety distance monitoring data for the head and shoulder area to obtain a comprehensive safety distance monitoring result.

[0125] Acquire safe distance monitoring data for the most prominent body part and routine safe distance monitoring data for the head and shoulder area. Integrate and merge these two types of monitoring data according to body parts to form a monitoring data set with a unified format. Based on the instantaneous distance values ​​of each body part in the monitoring data set, determine the safe distance monitoring status covering the most prominent body part and the head and shoulder area to obtain a comprehensive safe distance monitoring result.

[0126] In live-line working scenarios on power distribution networks, the comprehensive safe distance monitoring result is a complete monitoring record formed by fusing monitoring data from the most prominent part of the body with monitoring data from the head and shoulder area. The safe distance monitoring data for the most prominent part reflects the distance between the part of the body closest to the live conductor when the elbow is abducted and the surrounding live conductors, while the conventional safe distance monitoring data for the head and shoulder area reflects the distance between the worker's head and shoulders and the surrounding live conductors.

[0127] For example, when the two types of monitoring data are integrated according to human body parts, the source regions of the monitoring data of the most prominent part and the monitoring data of the head and shoulder area are marked respectively, so that each record in the monitoring data set can be clearly associated with a specific part of the human body.

[0128] In one possible implementation, the comprehensive safety distance monitoring results include the instantaneous distance between the most prominent part and the nearest charged object, the instantaneous distance between the head and shoulder area and the nearest charged object, and the location information of the charged object corresponding to each part. This comprehensive monitoring result covers the elbow abduction area and the head and shoulder area of ​​the worker. By fusing data from these areas and combining it with human posture estimation algorithms (such as skeletal keypoint tracking models), the safe distances to other limb parts such as arms and legs are inferred, achieving unified monitoring of human limb safety distances and avoiding monitoring blind spots caused by changes in limb posture.

[0129] If the technical solution of this application involves the collection, storage, use, processing, transmission, provision, disclosure, or deletion of personal information, the products using this technical solution have clearly and understandably informed the users of the personal information processing rules before processing personal information, and have obtained the individuals' voluntary consent in accordance with the law. If the technical solution of this application involves sensitive personal information (such as biometrics, religious beliefs, specific identities, medical and health information, financial accounts, and location tracking), the products using this solution have obtained the individuals' separate consent before processing sensitive personal information, and have also met the requirement of "express consent," ensuring that individuals make authorization decisions voluntarily based on full knowledge.

[0130] Specific implementation methods include, but are not limited to, the following: setting up clear and prominent signs at personal information collection devices such as cameras and sensors to inform relevant personnel that they have entered the scope of personal information collection and that their personal information will be collected and processed. If an individual voluntarily enters the collection scope after being informed, it is deemed that they have agreed to the collection of their personal information; or using obvious icons, text descriptions, or other means on the terminal device or system interface for personal information processing to inform them of the rules for personal information processing, and obtaining the individual's explicit authorization through interactive methods such as pop-up prompts, check confirmation boxes, or asking the individual to upload their personal information themselves.

[0131] The aforementioned personal information processing rules should include, but are not limited to, the name and contact information of the personal information processor, the specific purpose of personal information processing, the processing method, the types of personal information processed, the retention period, and the methods and procedures for individuals to exercise their relevant rights.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for full-process safety situation awareness during live-line working in power distribution networks, characterized in that: include: The real-time three-dimensional coordinates of each joint of the operator are obtained from the multi-joint skeletal tracking unit deployed on the insulated bucket truck. The real-time angle between the line connecting the shoulder joint and the elbow joint, and the line connecting the elbow joint and the wrist joint are extracted as the elbow abduction angle. Based on the real-time three-dimensional coordinates of each joint of the worker, combined with the preset skin layer thickness and contour expansion distance at each bone node, a set of three-dimensional coordinates of key contour points on the worker's body surface is generated. The original distance data of the surrounding charged bodies is collected by the omnidirectional laser ranging unit deployed on the insulated bucket truck. The original distance data of the surrounding charged bodies is fused with the three-dimensional coordinate set of key contour points on the body surface to extract the instantaneous distance set between each contour point of the human body and the surrounding charged bodies. The real-time elbow abduction angle is compared with a preset first threshold. When the elbow abduction angle exceeds the first threshold, a dynamic recognition process is triggered. After the dynamic recognition process is triggered, the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle are retrieved. The spatial boundary range of the area extending outward from the elbow joint and the subset of all contour points falling within the boundary are identified. The instantaneous distance value between each contour point in the contour point subset and the surrounding charged body is extracted. The contour point with the smallest instantaneous distance value is determined as the most prominent part of the human body at the current moment. The instantaneous distances between the most prominent part of the human body and each charged object are extracted from the instantaneous distance set to generate safe distance monitoring data for the most prominent part. At the same time, the instantaneous distances between each contour point of the head and shoulder area and the charged object are extracted from the instantaneous distance set to generate regular safe distance monitoring data for the head and shoulder area. The safety distance monitoring data for the most prominent parts is combined with the routine safety distance monitoring data for the head and shoulder areas to obtain a comprehensive safety distance monitoring result.

2. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, The process of acquiring the real-time three-dimensional coordinates of each joint of the operator, collected by the multi-joint skeletal tracking unit deployed on the insulated bucket truck, and extracting the real-time angles between the lines connecting the shoulder and elbow joints and the elbow and wrist joints as the elbow abduction angle includes: Obtain the real-time three-dimensional coordinates of the shoulder, elbow, and wrist joints of the operator; Generate the upper arm bone connection vector and the forearm bone connection vector; The vector connecting the upper arm bones and the vector connecting the forearm bones are processed to obtain the real-time angle between the two vectors, and this angle is determined as the elbow abduction angle at the current moment.

3. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, The process involves generating a set of three-dimensional coordinates for key contour points on the worker's body surface based on the real-time three-dimensional coordinates of each joint, combined with preset skin layer thickness and contour outward distance at each skeletal node. This set includes: Obtain the real-time three-dimensional coordinates of each joint of the operator; Retrieve the skin layer thickness and contour outward distance values ​​corresponding to each skeletal node from the preset human body surface parameter library; For each skeletal node, the radial direction pointing outward from the human body is determined based on the vertical direction of the line connecting adjacent bones. The skin layer thickness value and the contour outward expansion distance value are superimposed to obtain the comprehensive outward expansion amount. Using the three-dimensional coordinates of the skeletal node as a reference point, the coordinates are offset along the radial direction according to the comprehensive outward expansion to obtain the three-dimensional coordinates of the body surface contour point corresponding to the node. Traverse all skeletal nodes to generate the 3D coordinates of the body surface contour points corresponding to each node, forming a set of 3D coordinates of the key contour points on the worker's body surface.

4. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, The system collects raw distance data of surrounding charged bodies using an omnidirectional laser ranging unit deployed on an insulated bucket truck. This raw distance data is then fused with the three-dimensional coordinates of key contour points on the body surface to extract the instantaneous distance set between each contour point of the human body and the surrounding charged bodies, including: The omnidirectional laser ranging unit collects raw distance data and corresponding azimuth angles in all directions within the surrounding space. The reflection points on the surfaces of each charged body are converted into three-dimensional rectangular coordinates, thus obtaining the three-dimensional coordinates of the reflection points on the surfaces of the surrounding charged bodies. Spatial registration is performed by placing the three-dimensional coordinates of the key contour points on the body surface and the three-dimensional coordinates of the reflection points on the surface of each charged body in the same coordinate system. For each key contour point on the body surface, calculate the distance between the contour point and the reflection point on the surface of each charged body to obtain the instantaneous distance value between the contour point and the surrounding charged bodies. Traverse all contour points to form a set of instantaneous distances between each contour point of the human body and the surrounding charged bodies.

5. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, The step of comparing the real-time elbow abduction angle with a preset first threshold, and triggering a dynamic recognition process when the elbow abduction angle exceeds the first threshold, includes: Obtain the real-time elbow abduction angle at the current moment; Read the first threshold and compare the real-time elbow abduction angle with the first threshold. If the real-time elbow abduction angle exceeds the first threshold, it is determined that the current worker's elbow is in a state of large abduction, triggering the dynamic recognition process.

6. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, After the dynamic recognition process is triggered, the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle are retrieved. The spatial boundary range of the area extending outward from the elbow joint and the subset of all contour points falling within this boundary are identified. The instantaneous distance values ​​between each contour point in the subset and surrounding charged bodies are extracted, and the contour point with the smallest instantaneous distance value is determined as the most prominent part of the human body at the current moment, including: Retrieve the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle; Obtain the three-dimensional coordinates of the elbow joint as the starting point of the spatial boundary; Based on the angle between the line connecting the upper arm bones and the line connecting the forearm bones, the fan-shaped area extending outward from the elbow joint is determined as the spatial boundary. Traverse the three-dimensional coordinate set of the key contour points on the body surface, determine whether each contour point falls within the spatial boundary range, and include the contour points that fall within the boundary range into the contour point subset. Extract the instantaneous distance values ​​between each contour point in the subset of contour points and the surrounding charged bodies from the instantaneous distance set; The instantaneous distance values ​​of all contour points within the subset of contour points are numerically sorted. Extract the contour point with the smallest instantaneous distance value and determine that contour point as the most prominent part of the human body at the current moment.

7. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 6, characterized in that, The process of retrieving the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle, and identifying the spatial boundary range of the area extending outward from the elbow joint and the subset of contour points falling within that boundary, includes: Retrieve the three-dimensional coordinate set of key contour points on the body surface and the current elbow abduction angle; Obtain the three-dimensional coordinates of the elbow joint; Based on the elbow abduction angle and the direction of the line connecting the upper arm bones, determine the spatial boundary range extending outward from the elbow joint to the outside of the human body. Traverse the three-dimensional coordinate set of the key contour points on the body surface, filter out all contour points that fall within the spatial boundary range, and form a subset of contour points.

8. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, The method involves extracting the instantaneous distances between the most prominent part of the human body and each charged object from the instantaneous distance set to generate safe distance monitoring data for the most prominent part. Simultaneously, it extracts the instantaneous distances between each contour point in the head and shoulder region and charged objects from the instantaneous distance set to generate routine safe distance monitoring data for the head and shoulder region, including: Get the outline point number of the most prominent part of the human body at the current moment; Extract the instantaneous distance values ​​between the contour points of the most prominent part and each surrounding charged body, as well as the corresponding charged body position information, from the instantaneous distance set. Then, associate and combine the instantaneous distance values ​​and the charged body position information to generate safe distance monitoring data for the most prominent part. Based on the preset range of outline point numbers for the head and shoulder region, determine all outline points belonging to the head and shoulder region from the three-dimensional coordinate set of key outline points on the body surface. Extract the instantaneous distance values ​​between each contour point in the head and shoulder region and the surrounding charged bodies from the instantaneous distance set to generate routine safe distance monitoring data for the head and shoulder region; The safe distance monitoring data for the most prominent part is output in parallel with the regular safe distance monitoring data for the head and shoulder area.

9. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 8, characterized in that, The process of obtaining the contour point number of the most prominent part of the human body at the current moment, and extracting the instantaneous distance values ​​between the contour point of the most prominent part and each surrounding charged body, as well as the corresponding position information of the charged body, from the instantaneous distance set includes: Obtain the three-dimensional coordinates of the most prominent part of the human body at the current moment and the corresponding contour point number; Based on the contour point number, extract the instantaneous distance values ​​between the point and each surrounding charged body from the instantaneous distance set; Obtain the spatial coordinates of each charged body; The three-dimensional coordinates and instantaneous distance values ​​of the most prominent part are correlated with the spatial coordinates of each charged body to form safe distance monitoring data that includes the coordinates of the most prominent part, the position of the nearest charged body, and the instantaneous distance value.

10. The method for full-process safety situation awareness of live-line working in distribution networks according to claim 1, characterized in that, The method involves fusing safety distance monitoring data for the most prominent areas with routine safety distance monitoring data for the head and shoulder region to obtain a comprehensive safety distance monitoring result, including: Acquire safe distance monitoring data for the most prominent areas and routine safe distance monitoring data for the head and shoulder areas; The monitoring data is integrated and merged according to human body parts to form a monitoring data set with a unified format; Based on the instantaneous distance values ​​of each part in the monitoring data set, the safe distance monitoring status covering the most prominent part and the head and shoulder area is determined, and a comprehensive safe distance monitoring result is obtained.

Images