Substation personnel safety operation early warning system and method fusing UWB and imaging technology

Abstract

The invention discloses a substation personnel safety operation early warning system and method fusing UWB and imaging technology, and the system comprises the steps: a plurality of cameras are distributed in all regions in a substation, collect the image information in the substation in real time, and transmit the image data to a data fusion and analysis module; the UWB tag is configured to a person entering the transformer substation and periodically transmits UWB signals; the UWB base station receives the UWB signal and transmits the UWB signal to the data fusion and analysis module; the data fusion and analysis module carries out deep fusion on data acquired by the camera monitoring system and the UWB system; the weight evaluation module is used for comprehensively evaluating factors such as the safety set distance, the walking direction and the walking speed of the personnel according to a preset weight standard; and the early warning module triggers safety early warning of a corresponding level according to the safety risk score. The safety of transformer substation personnel is effectively guaranteed, and safety accidents caused by improper walking of the personnel are reduced.

Description

Technical Field

[0001] The invention relates to a transformer substation personnel safety operation early warning system and method integrating UWB and imaging technology, belonging to the technical field of transformer substation safety. Background Art

[0002] Substations are key components of the power system, with a large number of high-voltage equipment and complex electrical lines inside. Improper movement of personnel in substations may lead to safety accidents such as electric shock and misoperation, which seriously threaten the safety of personnel and the stable operation of the power grid. Traditional substation safety protection measures mainly include safety signs and fences, but these methods cannot effectively monitor and predict the real-time location and movement trajectory of personnel. Although there are some monitoring systems based on different technologies, they are insufficient in accuracy, real-time performance, and the function of predicting personnel movement.

[0003] At present, there are many problems that need to be solved in the existing technology. On the one hand, the traditional substation safety warning system is insufficient in accurately predicting the movement of people. It cannot accurately measure the risk level of people approaching dangerous areas, which greatly affects the effectiveness of safety assurance. On the other hand, among the existing solutions, there is no perfect safety prediction system based on UWB technology and camera image recognition technology that comprehensively sets multiple factors such as distance, walking direction and walking speed for personnel safety. The existing technical defects lead to a lack of comprehensiveness and accuracy in assessing personnel safety risks, and it is easy to miss key risk information. Summary of the invention

[0004] The purpose of the present invention is to provide a substation personnel safety operation warning system and method that integrates UWB and imaging technology. The system and method assign weights to factors related to personnel movement to accurately assess personnel safety risks, and then trigger appropriate warning measures in a timely manner to comprehensively protect the life safety of personnel in the substation and the stable operation of equipment.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A substation personnel safety operation warning system integrating UWB and imaging technology, including:

[0007] Camera monitoring system, UWB system, data fusion and analysis module, weight assessment module and early warning module;

[0008] The camera monitoring system: consists of multiple cameras distributed in various areas of the substation, which collect image information in the substation in real time and transmit the image data to the data fusion and analysis module;

[0009] The UWB system includes a UWB tag and a UWB base station. The UWB tag is configured for personnel entering the substation. Each tag has a unique identification code and periodically transmits a UWB signal. The UWB base station is used to receive the signal sent by the UWB tag and transmit the received signal data to the data fusion and analysis module.

[0010] The data fusion and analysis module receives image data from the camera monitoring system and signal data from the UWB system. The camera monitoring system processes the collected images in real time, identifies the personnel target, and then detects and locates the personnel, and obtains the three-dimensional position, walking speed and direction of the personnel. The UWB system determines the position coordinates, walking speed and direction of the personnel carrying the UWB tag. The data fusion processing deeply integrates the data obtained by the camera monitoring system and the UWB system.

[0011] The weight evaluation module receives data from the data fusion and analysis module and performs a comprehensive evaluation of factors such as the distance from the personnel to the safety setting, the walking direction, and the walking speed according to a preset weight standard;

[0012] The early warning module triggers a security early warning of a corresponding level when the security risk score output by the weight assessment module exceeds a preset threshold.

[0013] A method for early warning of safe operation of substation personnel integrating UWB and imaging technology comprises the following steps:

[0014] 1) Hardware deployment and data collection stage: Multiple cameras are distributed in various areas of the substation to collect image information in real time;

[0015] Equip personnel entering the substation with UWB tags, bind their basic information to the tag's identification code, and initialize the tag;

[0016] When the UWB tag enters the substation, the UWB tag periodically transmits signals and the UWB base station receives the signals;

[0017] 2) Data processing and analysis stage: Use the target detection algorithm in deep learning to process the images collected by the camera in real time, identify the human targets in the image, detect and locate the personnel, and obtain the three-dimensional position, walking speed and direction of the personnel who do not carry UWB tags; then perform behavioral analysis on the detected personnel images, use posture estimation technology to obtain the posture information of the personnel, and judge the walking, running, and staying status of the personnel; calculate the walking speed according to the position change of the personnel in the continuous image frames, and determine the walking direction by analyzing the body posture of the personnel and the feature points in the image;

[0018] For UWB systems, the time of flight (TOF) or time difference of arrival (TDOA) algorithm of the UWB signal is used to calculate the distance between the UWB tag and the base station, and then the position coordinates of the person are determined through the multilateral positioning method. Then, based on the person's position data at consecutive time points, the walking speed and direction of the person are calculated;

[0019] 3) During the risk assessment phase, weights are assigned to factors related to personnel movement to accurately assess personnel safety risks:

[0020] (1) Set the weight W of the distance between the person and the safety setting 1 :Determined according to the area division and equipment hazard level within the substation;

[0021] (2) Set the walking direction weight W 2 : It is determined by calculating the angle θ between the line connecting the personnel position and the geometric center of the dangerous area and the personnel walking direction vector;

[0022] (3) Set walking speed weight W 3 ;

[0023] (4) Risk assessment formula: S = W 1 ×D+W 2 ×DIR+W 3 ×V, where D is the normalized value of the distance from the person to the nearest danger zone boundary, d is the distance between the person and the boundary of the danger zone, and the value range of D is between 0 and 1, with a maximum distance of 100 meters; DIR is the normalized value of the degree to which the walking direction points to the danger zone. The value range of DIR is between 0 and 1, with 1 being completely pointing to the danger zone and 0 being completely away from it; V is the normalized value of walking speed. v max Take 2m / s, and the value range of v is between 0-1;

[0024] 4) Early warning trigger and response stage:

[0025] (1) Warning level classification: Different warning levels are classified according to security risk scores;

[0026] (2) Content of warning information sent: The warning information covers the location coordinates, behavior status, security risk score and risk factors of the personnel, so that the monitoring personnel can understand the situation and take appropriate measures.

[0027] The purpose of the present invention can also be further achieved by the following technical measures:

[0028] In the aforementioned substation personnel safety operation warning system integrating UWB and imaging technology, in the data fusion and analysis module, the data fusion processing process includes:

[0029] Unified data format: that is, the coordinate systems of the two systems are converted so that the three-dimensional position information of personnel based on image pixel coordinates obtained by the camera and the position coordinates of the UWB system based on the actual spatial coordinate system of the substation are in the same system. At the same time, the time stamp is used to synchronize the data in the time dimension to ensure accurate matching of personnel information at each moment.

[0030] Information association and integration: With personnel identification function, the image features of personnel identified by the camera are associated with the personnel identity corresponding to the UWB tag to ensure the consistency of personnel movement trajectory and risk status analysis;

[0031] When multi-source data is complementary and integrated, the location information is based on the UWB high-precision location and corrected and supplemented with the visual context information of the camera image; the speed and direction information are combined with the calculation results of the two systems. When normal, they are mutually verified. When there is a difference, weighted average is used or the reliable source is selected according to reliability.

[0032] In the stage of exception handling and quality assessment, a data quality detection mechanism is established to eliminate abnormal data caused by equipment failure or signal interference, and historical data and preset models are compared in real time to evaluate the quality of fusion results. When deviation or uncertainty increases, the detection and adjustment mechanism is automatically triggered to ensure data quality and the accuracy of risk assessment and early warning.

[0033] In the aforementioned early warning method for safe operation of substation personnel integrating UWB and imaging technology, in the risk assessment stage, for extremely dangerous areas of high-voltage equipment and large transformers, the personnel are set to a safe distance weight W. 1 Set to 50, for general passages and office areas, the distance weight W of personnel from the safety setting 1 Set to 20;

[0034] When 0°≤θ≤90°, the walking direction points to the dangerous area, and the walking direction weight W 2 Set to 30; when 90°≤θ≤180°, the walking direction is far away from the dangerous area, and the walking direction weight W 2 Set to 10;

[0035] When the walking speed exceeds 1.8m / s, the walking speed weight W 3 Set to 20, when the walking speed is between 1.3-1.8m / s, the walking speed weight W 3 Set to 10, when the walking speed is below 1.3m / s, the walking speed weight W 3 Set to 5.

[0036] The above-mentioned substation personnel safety operation warning method integrating UWB and imaging technology has the following warning level division method: when 70≤S, a high-level warning is triggered, and the strong flashing warning lights and high-decibel voice alarms throughout the station are activated, and an emergency alarm containing detailed personnel information and dangerous situations is sent to the monitoring center; when 40≤S<70, an intermediate warning is triggered, and the warning lights and voice prompts in the local area are activated, and the relevant personnel and risk information are displayed in the monitoring center; when 20≤S<40, a low-level warning is triggered, and the possible risk situations are only reminded to the personnel in the monitoring center.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] Through the high-resolution and wide dynamic range cameras in the camera monitoring system, image data can be collected from all areas of the substation. The deep learning target detection and posture estimation technology can be used to detect, locate and analyze the behavior of personnel. The three-dimensional position, walking speed and direction of personnel without UWB tags can be accurately obtained, providing key data for subsequent risk assessment, effectively filling the monitoring gap that exists only with UWB technology. The UWB system receives the tag signal from the base station for the personnel carrying the tag and calculates its position, speed and direction through the positioning algorithm, which complements the camera system data. The data fusion and analysis module integrates and unifies the personnel information obtained by the two, so that data from different sources can be analyzed collaboratively. The weight assessment module sets the weight of the personnel's distance from the safety setting, walking direction, and walking speed according to the degree of danger in different areas of the substation, and uses the risk assessment formula to accurately quantify the personnel safety risk score, avoiding the ambiguity of risk assessment in traditional monitoring. The early warning module triggers different levels of early warnings according to the risk score. When a high-level warning is triggered, an audible and visual alarm is sounded throughout the station and a detailed alarm is sent to the monitoring center. An intermediate warning activates a local warning and displays information. A low-level warning prompts the monitoring center, allowing monitoring personnel to quickly take measures based on the personnel location, behavioral status, risk score and risk factors in the early warning information, thereby effectively ensuring the safety of substation personnel and reducing safety accidents caused by improper movement of personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a structural diagram of the substation personnel safety operation warning system integrating UWB and imaging technology of the present invention;

[0040] Figure 2 It is a flow chart of the early warning method for safe operation of substation personnel integrating UWB and imaging technology of the present invention;

[0041] Figure 3 It is a schematic diagram of signal transmission of UWB system;

[0042] Figure 4 It is a schematic diagram of the camera monitoring system. DETAILED DESCRIPTION

[0043] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0044] The present invention integrates UWB technology and camera image recognition technology to create a safety prediction system that can cover all personnel (regardless of whether they carry UWB tags). Camera image recognition and analysis technology is used to capture the location, walking direction, speed and other information of personnel without UWB tags, and UWB technology is combined to obtain relevant data of personnel carrying tags, and then a scientific and reasonable weight system is established based on key factors such as the distance from the personnel to the safety setting, walking direction, and walking speed to comprehensively judge the safety risk level of the personnel. Accurate early warning information is triggered according to the risk assessment results, covering different levels of early warning from low to high, and the early warning content is detailed and accurate, including the location of the personnel, behavioral status, safety risk score and specific risk factors, etc., so that monitoring personnel can grasp the dynamics and potential dangers of personnel in a timely manner, take targeted measures, and effectively prevent safety accidents caused by improper movement of personnel in high-risk environments such as substations, and ensure the safe and stable operation of substations.

[0045] 1. System Structure

[0046] like Figure 1 As shown, the substation personnel safety operation warning system integrating UWB and imaging technology of the present invention includes a camera monitoring system, a UWB system (including a UWB tag and a UWB base station), a data fusion and analysis module, a weight evaluation module and an early warning module:

[0047] 1. Camera monitoring system: It consists of multiple high-resolution cameras with wide dynamic range. These cameras are widely distributed in various areas of the substation, including high-voltage equipment areas, channels, operating rooms, etc., to ensure that there are no blind spots in monitoring. The camera is responsible for real-time acquisition of image information in the substation and transmits the image data to the data fusion and analysis module.

[0048] 2. UWB system: UWB tags are used to distribute to some or all personnel entering the substation. Each tag has a unique identification code and can periodically transmit UWB signals. UWB base stations are distributed at key locations in the substation to receive signals from UWB tags and transmit the received signal data to the data fusion and analysis module.

[0049] 3. Data fusion and analysis module: This module receives image data from the camera monitoring system and signal data from the UWB system. For the camera monitoring system, it is mainly used to monitor personnel who do not carry UWB tags. Through multiple high-resolution and wide dynamic range cameras widely distributed in the substation, the target detection algorithm in deep learning is used to process the collected images in real time, identify personnel targets, and then detect and locate personnel, and obtain the three-dimensional position, walking speed and direction of personnel who do not carry UWB tags. Its advantage is that it can intuitively obtain the image information of personnel, which is suitable for monitoring personnel who are not equipped with specific equipment. For the UWB system, for personnel carrying UWB tags, UWB tags periodically transmit signals, and UWB base stations distributed at key locations of the substation receive signals, and use mature positioning algorithms (such as time of flight TOF time difference of arrival TDOA, etc.), combined with the time information of multiple base stations receiving signals, calculate the distance between the UWB tag and the base station, determine the position coordinates of the personnel through multilateral positioning methods, and then calculate the walking speed and direction of the personnel based on the personnel position data at continuous time points. The system has high accuracy in locating personnel carrying tags. The two complement each other and jointly realize comprehensive monitoring of all personnel in the substation (regardless of whether they carry UWB tags). The camera monitoring system makes up for the shortcomings of the UWB system in monitoring personnel without tags, while the UWB system provides accurate positioning information for personnel carrying tags. The combination of the two can more comprehensively and accurately grasp the dynamic situation of personnel in the substation. Data fusion processing is committed to deeply integrating the data obtained by the camera monitoring system and the UWB system. Its specific process covers the unification of data formats, that is, converting the coordinate systems of the two systems so that the three-dimensional position information of personnel based on image pixel coordinates, etc. obtained by the camera and the position coordinates of the UWB system based on the actual spatial coordinate system of the substation are in the same system. At the same time, the timestamp is used to synchronize the data in the time dimension to ensure accurate matching of personnel information at all times; in terms of information association and integration, if the personnel identity recognition function is available, the image features of the personnel recognized by the camera can be associated with the personnel identity corresponding to the UWB tag to ensure the consistency of personnel movement trajectory and risk status analysis. When multi-source data are complementary and fused, for location information, the high-precision positioning of UWB is used. The system takes the position as the basis, and uses the visual context information of the camera image to correct and supplement it. The speed and direction information are integrated with the calculation results of the two systems. When normal, they verify each other. When there is a difference, the weighted average is used or the trusted source is selected according to reliability. In the abnormal processing and quality assessment stage, by establishing a data quality detection mechanism, abnormal data caused by equipment failure or signal interference, such as large jumps in the position of the camera personnel and abnormally high UWB speed, are eliminated, and the quality of the fusion results is evaluated by real-time comparison of historical data and preset models. When the deviation or uncertainty increases, the detection adjustment mechanism is automatically triggered, such as increasing the image acquisition frequency or recalibrating the UWB base station, to ensure the accuracy of data quality and risk assessment and early warning.

[0050] 4. Weight assessment module: Based on the preset weight standards, a comprehensive assessment is conducted on key factors such as the distance from the safety setting, walking direction, walking speed, etc. These weight values ​​are set based on factors such as the degree of danger in different areas of the substation and equipment operation requirements, and the safety risk score of the personnel is obtained through a specific calculation model.

[0051] 5. Early warning module: According to the security risk score output by the weight assessment module, when the score exceeds the preset threshold, the corresponding level of security warning is triggered, including but not limited to sound and light alarms, sending early warning information to the monitoring center, etc.

[0052] 2. Work steps flow, such as Figure 2 , 3 , as shown in 4:

[0053] 1. Hardware deployment stage: For the camera monitoring system, the camera monitoring system is rationally planned and installed in the substation, including determining the number, location, angle, etc. of the cameras, to ensure full coverage of all areas of the substation and no blind spots in monitoring. The network configuration of the camera is carried out, and its resolution, frame rate and other parameters are set to ensure that the image data can be transmitted stably and clearly; for the UWB system, according to the scale and complexity of the substation, UWB base stations are installed at key locations to ensure that the signal coverage between base stations can cover all areas where people may appear, and the UWB base stations are calibrated and parameterized, such as setting the signal transmission power, receiving sensitivity, working frequency band, etc. UWB tags are equipped for personnel entering the substation, and the basic information of the personnel is bound to the identification code of the tag, and the tag is initialized at the same time.

[0054] 2. Data collection stage: For the camera monitoring system, when the system is started, the camera continuously collects image information in the substation and transmits the image data to the data fusion and analysis module in real time; for the UWB system, when the UWB tag enters the substation, it starts to transmit signals periodically, and the UWB base station receives these signals and transmits the data to the data fusion and analysis module.

[0055] 3. Data processing and analysis stage: For camera monitoring systems, camera image recognition technology uses the target detection algorithm in deep learning to process the images collected by the camera in real time, identify the human targets in the image, and realize the detection and positioning of the personnel. It can accurately obtain the three-dimensional position, walking speed and direction of the personnel who do not carry UWB tags. After that, the detected personnel images are further analyzed for behavior, and the posture estimation technology is used to obtain the posture information of the personnel, so as to judge the walking, running, staying and other behavioral states of the personnel. The walking speed is calculated according to the position change of the personnel in the continuous image frames, and the walking direction is determined by analyzing the body posture of the personnel and the feature points in the image. For example, by analyzing the image position change of the personnel within a certain time interval, the displacement components in different directions are calculated, and then the walking speed and direction vector are obtained; for UWB systems, the flight time (TOF) or arrival time difference (TDOA) of the UWB signal is used. The time information of the received signals of multiple UWB base stations is combined to calculate the distance between the UWB tag and the base station, and then the position coordinates of the personnel are determined by the multilateral positioning method, and then the walking speed and direction of the personnel are calculated based on the personnel position data at continuous time points.

[0056] 4. Risk assessment stage:

[0057] 1) Weight W of the distance between personnel and safety setting 1 :Determined based on the area division and equipment hazard level within the substation. For extremely dangerous areas such as those close to high-voltage equipment and large transformers, a slight careless approach by personnel may cause serious accidents. In these areas, the distance between personnel and the boundary of the area has a huge impact on safety risks, and this weight can be set to 50. For general passages, office areas and other areas far away from major dangerous equipment, the change in the distance between personnel and the safety setting has little impact on safety risks, and the weight can be set to 20.

[0058] 2) Walking direction weight W 2 :The walking direction of the personnel also has a key impact on the assessed risk of the dangerous area, which is determined by calculating the angle θ between the line connecting the personnel position and the geometric center of the dangerous area and the walking direction vector of the personnel. When 0°≤θ≤90°, the walking direction points to the vicinity of the dangerous area, that is, the personnel is approaching the dangerous area, and this weight can be set to 30. When 90°≤θ≤180°, the walking direction is far away from the dangerous area, and the risk is relatively low, and this weight can be set to 10.

[0059] 3) Walking speed weight W 3:The impact of personnel walking speed on safety risks is different in different situations. In the substation, if the personnel walk fast, especially when approaching dangerous areas, once an accident occurs, it may be too late to react, resulting in a serious accident. Therefore, when the personnel walking speed exceeds 1.8m / s, when approaching dangerous areas, the safety risk will increase greatly, and the weight can be set to 20. When the personnel walking speed is between 1.3-1.8m / s, the personnel speed is slightly reduced, but there is still a certain degree of risk. At this time, the weight can be set to 10. When the personnel walking speed is below 1.3m / s, it can be regarded as a normal walking speed. At this time, the impact of walking speed on safety risks is relatively small, and the weight can be set to 5.

[0060] 4) Risk assessment formula: S = W 1 ×D+W 2 ×DIR+W 3 ×V, where D is the normalized value of the distance between the person and the nearest danger zone boundary ( d refers to the distance between the person and the boundary of the danger zone. The value of D ranges from 0 to 1, with a maximum distance of 100 meters. The closer the distance, the larger the value. DIR is the normalized value of the degree to which the walking direction points to the danger zone ( DIR ranges from 0 to 1, with 1 being completely pointing to the danger zone and 0 being completely away from it), and V is the normalized value of walking speed ( v max It can be 2m / s, and the value range of v is between 0-1. The higher the speed, the larger the value).

[0061] 5. Early warning trigger and response stage

[0062] 1) Warning level classification: Different warning levels are divided according to the safety risk score. When 70≤S, a high-level warning is triggered, and the strong flashing warning lights and high-decibel voice alarms throughout the station are activated, and an emergency alarm containing detailed personnel information and dangerous situations is sent to the monitoring center; when 40≤S<70, a medium-level warning is triggered, and the warning lights and voice prompts in the local area are activated, and the relevant personnel and risk information are displayed in the monitoring center; when 20≤S<40, a low-level warning is triggered, and only the monitoring center reminds people of possible risks.

[0063] 2) Warning information content: The warning information covers the person’s location coordinates (obtained through camera image recognition or UWB positioning), behavior status (such as walking, running, staying, etc.), safety risk score and major risk factors (such as being too close to a dangerous device, walking towards a dangerous area, walking too fast, etc.), so that monitoring personnel can quickly and accurately understand the situation and take appropriate measures.

[0064] In addition to the above embodiments, the present invention may also have other implementation modes. Any technical solutions formed by equivalent replacement or equivalent transformation shall fall within the protection scope required by the present invention.

Claims

1. A substation personnel safety operation warning system integrating UWB and imaging technology, characterized in that: include: Camera monitoring system, UWB system, data fusion and analysis module, weight assessment module and early warning module; The camera monitoring system: consists of multiple cameras distributed in various areas of the substation, which collect image information in the substation in real time and transmit the image data to the data fusion and analysis module; The UWB system includes a UWB tag and a UWB base station. The UWB tag is configured for personnel entering the substation. Each tag has a unique identification code and periodically transmits a UWB signal. The UWB base station is used to receive the signal sent by the UWB tag and transmit the received signal data to the data fusion and analysis module. The data fusion and analysis module receives image data from the camera monitoring system and signal data from the UWB system. The camera monitoring system processes the collected images in real time, identifies the personnel target, and then detects and locates the personnel, and obtains the three-dimensional position, walking speed and direction of the personnel. The UWB system determines the position coordinates, walking speed and direction of the personnel carrying the UWB tag. The data fusion processing deeply integrates the data obtained by the camera monitoring system and the UWB system. The weight evaluation module receives data from the data fusion and analysis module and performs a comprehensive evaluation of factors such as the distance from the personnel to the safety setting, the walking direction, and the walking speed according to a preset weight standard; The early warning module triggers a security early warning of a corresponding level when the security risk score output by the weight assessment module exceeds a preset threshold.

2. The substation personnel safety operation warning system integrating UWB and imaging technology as claimed in claim 1 is characterized in that: In the data fusion and analysis module, the process of data fusion processing includes: Unified data format: that is, the coordinate systems of the two systems are converted so that the three-dimensional position information of personnel based on image pixel coordinates obtained by the camera and the position coordinates of the UWB system based on the actual spatial coordinate system of the substation are in the same system. At the same time, the time stamp is used to synchronize the data in the time dimension to ensure accurate matching of personnel information at each moment. Information association and integration: With personnel identification function, the image features of personnel identified by the camera are associated with the personnel identity corresponding to the UWB tag to ensure the consistency of personnel movement trajectory and risk status analysis; When multi-source data is complementary and integrated, the location information is based on the UWB high-precision location and corrected and supplemented with the visual context information of the camera image; the speed and direction information are combined with the calculation results of the two systems. When normal, they are mutually verified. When there is a difference, weighted average is used or the reliable source is selected according to reliability. In the stage of exception handling and quality assessment, a data quality detection mechanism is established to eliminate abnormal data caused by equipment failure or signal interference, and historical data and preset models are compared in real time to evaluate the quality of fusion results. When deviation or uncertainty increases, the detection and adjustment mechanism is automatically triggered to ensure data quality and the accuracy of risk assessment and early warning.

3. A method for early warning of safe operation of substation personnel integrating UWB and imaging technology, characterized in that: The following steps are involved: 1) Hardware deployment and data collection stage: Multiple cameras are distributed in various areas of the substation to collect image information in real time; Equip personnel entering the substation with UWB tags, bind their basic information to the tag's identification code, and initialize the tag; When the UWB tag enters the substation, the UWB tag periodically transmits signals and the UWB base station receives the signals; 2) Data processing and analysis stage: Use the target detection algorithm in deep learning to process the images collected by the camera in real time, identify the human targets in the image, detect and locate the personnel, and obtain the three-dimensional position, walking speed and direction of the personnel who do not carry UWB tags; then perform behavioral analysis on the detected personnel images, use posture estimation technology to obtain the posture information of the personnel, and judge the walking, running, and staying status of the personnel; calculate the walking speed according to the position change of the personnel in the continuous image frames, and determine the walking direction by analyzing the body posture of the personnel and the feature points in the image; For UWB systems, the time of flight (TOF) or time difference of arrival (TDOA) algorithm of the UWB signal is used to calculate the distance between the UWB tag and the base station, and then the position coordinates of the person are determined through the multilateral positioning method. Then, based on the person's position data at consecutive time points, the walking speed and direction of the person are calculated; 3) During the risk assessment phase, weights are assigned to factors related to personnel movement to accurately assess personnel safety risks: (1) Setting the weight of the distance between personnel and safety setting: determined according to the area division within the substation and the danger level of the equipment; (2) Setting walking direction weight: This is determined by calculating the angle between the line connecting the personnel position and the geometric center of the dangerous area and the personnel walking direction vector; (3) Setting walking speed weight; (4) Risk assessment formula: , where is the normalized value of the distance between the person and the nearest danger zone boundary, , is the distance between the person and the danger zone boundary, ranging from 0 to 1, with a maximum distance of 100 meters; is the normalized value of the degree to which the walking direction points to the danger zone, , ranging from 0 to 1, with 1 being completely pointing to the danger zone and 0 being completely away from it; is the normalized value of the walking speed, , taking 2 m / s, ranging from 0 to 1; 4) Early warning trigger and response stage: (1) Warning level classification: Different warning levels are classified according to security risk scores; (2) Content of warning information sent: The warning information covers the location coordinates, behavior status, security risk score and risk factors of the personnel, so that the monitoring personnel can understand the situation and take appropriate measures.

4. The method for early warning of safe operation of substation personnel integrating UWB and imaging technology as claimed in claim 3 is characterized in that: In the risk assessment stage, for extremely dangerous areas such as high-voltage equipment and large transformers, the weight W1 of the distance from the safety setting is set to 50, and for general passages and office areas, the weight W1 of the distance from the safety setting is set to 20; When 0°≤θ≤90°, the walking direction points to the dangerous area, and the walking direction weight W2 is set to 30; when 90°≤θ≤180°, the walking direction is away from the dangerous area, and the walking direction weight W2 is set to 10; When a person's walking speed exceeds 1.8 m / s, the walking speed weight W3 is set to 20; when the person's walking speed is between 1.3-1.8 m / s, the walking speed weight W3 is set to 10; when the person's walking speed is below 1.3 m / s, the walking speed weight W3 is set to 5.

5. The method for early warning of safe operation of substation personnel integrating UWB and imaging technology as claimed in claim 3 is characterized in that: The warning level classification method is as follows: when 70 ≤ S, a high-level warning is triggered, strong flashing warning lights and high-decibel voice alarms are activated throughout the station, and an emergency alarm containing detailed personnel information and dangerous situations is sent to the monitoring center; When 40≤S<70, an intermediate warning is triggered, the warning lights and voice prompts in the local area are activated, and the relevant personnel and risk information are displayed in the monitoring center; when 20≤S<40, a low-level warning is triggered, and people are only reminded of possible risk situations in the monitoring center.

Images