Intelligent safety management early warning system for electric power safety tools

Abstract

The invention discloses an intelligent safety management early warning system for an electric power safety tool, and relates to electric power operation early warning. Physical sign risk assessment based on an artificial intelligence algorithm model is combined, an assessment level is established, and the physical sign risk of an operator is comprehensively judged according to an assessment result obtained by the algorithm model; according to the scheme, the algorithm model is combined for data evaluation, so that the evaluation result is more accurate; and the risk level of the region is judged through the density, risk degree and other characteristics of historical data of the region positioned by the device, and the operator is prompted according to the risk level, so that initial early warning can be performed before operation.

Description

Technical Field

[0001] The present invention relates to the field of early warning of power distribution network for electric power operation, and in particular to an intelligent safety management early warning system for electric power safety tools. Background Art

[0002] Electricity inspection is a key safety inspection task that aims to ensure the safe operation of electricity facilities and power grids. This work not only involves inspecting power lines, meters, distribution equipment, etc. to identify and repair potential safety hazards, such as aging lines, meter failures, or improper wiring, but also includes the management and monitoring of distribution networks. By implementing smart distribution network technologies such as smart metering, demand response management, and the integration of distributed energy resources, grid operation can be optimized, failures can be prevented, and the reliability of power supply can be improved.

[0003] The purpose of power inspection is to ensure the stability of the power system and the safety of customers' power use. It includes the use of advanced technologies, such as drones to inspect high-altitude power lines, and the application of artificial intelligence and big data analysis to predict and prevent power grid failures. The inspection work usually includes several key steps. First, inspectors will conduct a visual inspection of the power lines, looking for obvious signs of damage or wear. Then, they will use professional tools to detect the current and voltage of the lines to assess the operating status of the lines. In addition, it also includes checking the meters and distribution boxes to ensure that the measurement is accurate and has not been illegally tampered with.

[0004] Theft of electricity greatly increases the risk of electric shock and fire. Since the theft of electricity is diverse and covert, it is difficult to inspect electricity usage. If workers fail to accurately identify tampered or illegally wired equipment, it may lead to negligence, resulting in personal electric shock. In addition, illegal connection or tampering of power lines may also cause circuit overload and cause fire accidents.

[0005] The shortcomings of the existing technology include: (1) Workers fail to accurately identify live wires and equipment, which can easily lead to negligence and cause electric shock accidents; (2) The current electric proximity warning equipment uses electric proximity induction sensors for warning, which has a small warning range and a single effect, and cannot fully guarantee the personal safety of power workers; (3) Existing early warning equipment is large in size, not suitable for power operation environments, and is complex to operate. Summary of the invention

[0006] The present invention and the application scheme aim to provide a system for early warning of power operations based on biological signs. The system provides early warning and protection for the personal safety of power operators in the distribution network through the positioning of working devices and the collection and judgment of vital sign information.

[0007] Among them, the intelligent safety management and early warning system for electric power safety tools includes a biological sign collection device and a safety early warning terminal. The biological sign collection device is integrated with a sensor group for collecting biological signs and positioning. The safety early warning terminal is connected to the biological sign collection device through cloud collaboration, wherein: The safety warning terminal is integrated with a risk warning subsystem, which is used to perform a sign risk assessment based on an artificial intelligence algorithm model on the biological sign information collected by the sensor group, establish an assessment level, and make a comprehensive judgment and warning on the sign risk of the operator according to the assessment results obtained by the algorithm model; The safety warning terminal is integrated with a risk prediction subsystem, which is used to obtain the historical danger characteristics of the positioning area based on the positioning information collected by the sensor group, determine the risk level of the area according to the number of historical danger characteristics, and send a warning signal based on the regional risk level.

[0008] Furthermore, the risk warning subsystem specifically includes: A biological sign data acquisition module is used to acquire the biological sign data collected by the sensor group and normalize the data; A feature extraction module is used to extract features from the biometric data after data processing, and the extracted features include at least heart rate, blood pressure, body temperature, and fatigue index; A risk level establishment module, used to assign a risk level to the extracted features, wherein the risk level is at least one of low risk, medium risk, and high risk; The decision tree building module is used to traverse all features and segmentation points through the CART algorithm, and to perform data segmentation according to the best segmentation method of the Gini impurity criterion, and to recursively segment the subsets generated by each segmentation until the preset conditions are met.

[0009] Furthermore, the specific process of data segmentation in the decision tree construction module is as follows: Evaluate the split point, traverse all features and split points, and select the split method that minimizes the Gini impurity, where Representation dataset Middle The proportion of classes Represents a dataset composed of features.

[0010] Furthermore, the risk prediction subsystem specifically includes: A historical area model grade assessment establishment module is used to classify the risk grade of each area in the historical area model according to the historical hazard characteristics; A positioning data acquisition module is used to acquire the positioning data collected by the sensor group, and acquire the corresponding area of ​​the positioning area and the historical area model according to the positioning data; The regional risk level warning module is used to issue warnings based on the risk level of the area where the positioning data is located.

[0011] Furthermore, the hazard characteristics include at least the total number of historical electricity thefts in the area, the historical density of electricity theft in the area, and the historical hazard level of distributors in the area; the total number of historical electricity thefts in the area indicates the total number of times the distributors in the area have been stolen, the historical density of electricity theft in the area indicates the frequency of the theft of electricity from the distributors in the area within a specified time period, the historical hazard level of distributors in the area indicates the degree of strain on the distributors in the area, and the historical hazard level of distributors in the area is determined by the length of time the distributors have not been repaired / updated.

[0012] Furthermore, the historical regional model level evaluation establishment module specifically includes: The risk scoring unit is used to perform risk scoring on the total number of electricity thefts in the region, the density of electricity theft in the region, and the danger level of the historical electricity theft in the region through sensitivity analysis; A risk level classification unit, used to classify the area into different risk levels based on the score, wherein the risk levels are low risk area, medium risk area and high risk area; Wherein, the risk scoring unit includes: The risk scoring model establishment subunit is used to establish a risk scoring model based on the total number of electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region; A Monte Carlo simulation subunit, used for generating parameter combination samples according to Monte Carlo simulation; The risk scoring subunit is used to calculate the corresponding risk score for the generated parameter combination sample according to the risk scoring model.

[0013] Furthermore, the risk level classification unit includes: The variance Gini coefficient sub-unit is used to calculate the variance Gini coefficient corresponding to the total number of electricity thefts in the region, the density of electricity theft in the region, and the danger level of the historical electricity theft in the region; A threshold setting subunit is used to set the thresholds corresponding to the total number of historical electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region according to the calculated variance Gini coefficient; The risk level classification subunit is used to classify the risk level according to the set threshold value combined with the risk score calculated by the risk scoring subunit.

[0014] Furthermore, in the variance Gini coefficient subunit, the calculation process of the variance Gini coefficient corresponding to the total number of historical electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region is as follows: ; ; Among them, the represents the risk score, represents the total variance of the risk score, Represents conditional expectation The variance of When the value of The degree of change of the mean value, M represents the parameter The total number of values, j represents the total number of historical power thefts in the area The index of the value of Representation parameters The jth specific value of Representation parameters equal When the risk score The expected value of Represents risk score The average value, They respectively represent the total number of historical electricity thefts in the region or the historical density of electricity theft in the region or the historical danger level of the distribution device in the region.

[0015] Furthermore, the warning of the risk warning subsystem is: the risk warning subsystem issues warning information to operators with high biological sign risk levels through biological sign collection equipment.

[0016] Furthermore, the warning of the risk prediction subsystem is: the risk warning subsystem uses biological sign collection equipment to provide warning information to operators with high risk levels in the area.

[0017] The beneficial effects of the invention are: (1) The present invention performs a physical sign risk assessment based on an artificial intelligence algorithm model on the physical sign information, establishes an assessment level, and makes a comprehensive judgment on the physical sign risk of the operator based on the assessment results obtained by the algorithm model. This solution combines the algorithm model to perform data assessment, making the assessment results more accurate; (2) The present invention aims at the situation where there are many cases of electricity theft and the power distribution environment is poor. By collecting regional historical data sets, a regional risk prediction system is established based on the regional historical data sets. According to the density, danger level and other characteristics of the historical data of the area where the device is located, the risk level of the area is judged, and the operating personnel are prompted according to the risk level. The process is implemented before the operation starts, and the purpose is to provide an initial warning of the danger level of the operation, so as to increase the crisis level of the operating personnel or back-end management personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a system structure diagram of the intelligent safety management and early warning system for electric power safety tools provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention. It should be noted that relational terms such as the terms "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0022] Moreover, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a list of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus. In the absence of more restrictions, an element defined by the phrase "comprises a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0023] The features and performance of the present invention are further described in detail below in conjunction with the embodiments.

[0024] like Figure 1 As shown, the intelligent safety management and early warning system for electric power safety tools includes a biological sign collection device and a safety early warning terminal. The biological sign collection device is integrated with a sensor group for collecting biological signs and positioning. The safety early warning terminal is connected to the biological sign collection device through cloud collaboration, wherein: The safety warning terminal is integrated with a risk warning subsystem, which is used to perform a sign risk assessment based on an artificial intelligence algorithm model on the biological sign information collected by the sensor group, establish an assessment level, and make a comprehensive judgment and warning on the sign risk of the operator according to the assessment results obtained by the algorithm model; The safety warning terminal is integrated with a risk prediction subsystem, which is used to obtain the historical danger characteristics of the positioning area based on the positioning information collected by the sensor group, determine the risk level of the area according to the number of historical danger characteristics, and send a warning signal based on the regional risk level.

[0025] Further, as a preferred specific implementation of this embodiment, a bio-signs collection bracelet for power distribution network operation is proposed. The bracelet has a high-sensitivity near-electric induction sensor integrated inside to detect the distance between the on-site operator and the charged object. When the distance to the charged object is less than the minimum distance specified in the "National Electricity Supply and Use Rules", the internal buzzer and vibrator will issue an anti-electric shock warning to prevent the operator from being electrocuted, thereby protecting personal safety. The bracelet uses two wireless communication modules, namely the Bluetooth communication module inside the bracelet and the GPRS communication module inside the mobile phone. According to the requirements of GB 26860-2011 "Electric Power Safety Work Regulations", a dedicated guardian is required at the power operation site to monitor the safety of the members of the operation team. Therefore, the Bluetooth communication function is integrated into the bracelet, and a monitoring platform APP is designed and developed and installed on the mobile phone of the dedicated guardian. The sensor on the bracelet uploads the collected information to the APP to facilitate the monitoring work of the dedicated personnel. The mobile phone returns the biological information (such as heart rate, blood pressure) and on-site work progress information of the on-site workers to the background terminal through the GPRS wireless communication module, so that the background can make unified management of the biological indicators of the workers to determine whether the workers can complete the work and the progress of the work project.

[0026] Furthermore, the risk warning subsystem specifically includes: A biological sign data acquisition module is used to acquire the biological sign data collected by the sensor group and normalize the data; A feature extraction module is used to extract features from the biometric data after data processing, and the extracted features include at least heart rate, blood pressure, body temperature, and fatigue index; A risk level establishment module, used to assign a risk level to the extracted features, wherein the risk level is at least one of low risk, medium risk, and high risk; The decision tree building module is used to traverse all features and segmentation points through the CART algorithm, and to perform data segmentation according to the best segmentation method of the Gini impurity criterion, and to recursively segment the subsets generated by each segmentation until the preset conditions are met.

[0027] Specifically, the above embodiment monitors the biological sign data of the operator in real time, such as heart rate, blood pressure, etc., and uses an integrated artificial intelligence algorithm model to analyze these data to assess the health and safety risks of the operator. The system will establish a risk level based on the evaluation results of the model. If the vital sign data is found to be abnormal or reaches a preset risk threshold, the system will issue an early warning to remind the operator or manager to take corresponding measures.

[0028] Furthermore, the specific process of data segmentation in the decision tree construction module is as follows: Evaluate the split point, traverse all features and split points, and select the split method that minimizes the Gini impurity, where Representation dataset Middle The proportion of classes Represents a dataset composed of features.

[0029] Furthermore, the risk prediction subsystem specifically includes: A historical area model grade assessment establishment module is used to classify the risk grade of each area in the historical area model according to the historical hazard characteristics; A positioning data acquisition module is used to acquire the positioning data collected by the sensor group, and acquire the corresponding area of ​​the positioning area and the historical area model according to the positioning data; The regional risk level warning module is used to issue warnings based on the risk level of the area where the positioning data is located.

[0030] Specifically, the above embodiment predicts the risk level of the operation area by analyzing the operator location information collected by the sensor group and combining it with the regional hazard feature data in the historical database, such as the total number of historical power theft events in the area, the power theft density, and the hazard level of the distributor, etc. The system determines the risk level of the operation area based on the number and severity of these historical data features, and sends an early warning signal accordingly to provide risk warning alerts to operators to help them avoid potential safety threats.

[0031] Furthermore, the hazard characteristics include at least the total number of historical electricity thefts in the area, the historical density of electricity theft in the area, and the historical hazard level of distributors in the area; the total number of historical electricity thefts in the area indicates the total number of times the distributors in the area have been stolen, the historical density of electricity theft in the area indicates the frequency of the theft of electricity from the distributors in the area within a specified time period, the historical hazard level of distributors in the area indicates the degree of strain on the distributors in the area, and the historical hazard level of distributors in the area is determined by the length of time the distributors have not been repaired / updated.

[0032] Furthermore, the historical regional model level evaluation establishment module specifically includes: The risk scoring unit is used to perform risk scoring on the total number of electricity thefts in the region, the density of electricity theft in the region, and the danger level of the historical electricity theft in the region through sensitivity analysis; A risk level classification unit, used to classify the area into different risk levels based on the score, wherein the risk levels are low risk area, medium risk area and high risk area; Wherein, the risk scoring unit includes: The risk scoring model establishment subunit is used to establish a risk scoring model based on the total number of electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region; A Monte Carlo simulation subunit, used for generating parameter combination samples according to Monte Carlo simulation; The risk scoring subunit is used to calculate the corresponding risk score for the generated parameter combination sample according to the risk scoring model.

[0033] Specifically, the above embodiment performs quantitative analysis on the historical electricity theft data in the region to generate a risk score. It includes two subunits: a risk score model establishment subunit and a Monte Carlo simulation subunit. The risk score model establishment subunit is used to build a mathematical model that assesses risk based on parameters such as the total number of historical electricity thefts, the density of electricity thefts, and the danger level of the distributor. The Monte Carlo simulation subunit is used to generate a large number of parameter combination samples and assess the uncertainty and sensitivity of the risk score by simulating the changes in these parameters.

[0034] Furthermore, the Monte Carlo simulation process randomly generates a large number of parameter combinations, and each parameter takes a random value within its predefined range; that is: , where T, F, and D correspond to the total number of historical electricity thefts in the distribution network, the historical electricity theft density in the region, and the historical danger level of the distribution network in the region, respectively.

[0035] Furthermore, the risk level classification unit includes: The variance Gini coefficient sub-unit is used to calculate the variance Gini coefficient corresponding to the total number of electricity thefts in the region, the density of electricity theft in the region, and the danger level of the historical electricity theft in the region; A threshold setting subunit is used to set the thresholds corresponding to the total number of historical electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region according to the calculated variance Gini coefficient; The risk level classification subunit is used to classify the risk level according to the set threshold value combined with the risk score calculated by the risk scoring subunit.

[0036] Specifically, the above embodiment calculates the risk score based on the historical electricity theft data in the region. The model will consider multiple factors, such as the total number of historical electricity thefts, the density of electricity thefts, and the dangerousness of the distributor, and determine how these factors jointly affect the risk level of the region through statistical and machine learning methods, and establish a calculation formula for the risk score accordingly.

[0037] Furthermore, in the variance Gini coefficient subunit, the calculation process of the variance Gini coefficient corresponding to the total number of historical electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region is as follows: ; ; Among them, the represents the risk score, represents the total variance of the risk score, Represents conditional expectation The variance of When the value of The degree of change of the mean value, M represents the parameter The total number of values, j represents the total number of historical power thefts in the area The index of the value of Representation parameters The jth specific value of Representation parameters equal When the risk score The expected value of Represents risk score The average value, They respectively represent the total number of historical electricity thefts in the region or the historical density of electricity theft in the region or the historical danger level of the distribution device in the region.

[0038] Furthermore, the warning of the risk warning subsystem is: the risk warning subsystem issues warning information to operators with high biological sign risk levels through biological sign collection equipment.

[0039] Furthermore, the warning of the risk prediction subsystem is: the risk warning subsystem uses biological sign collection equipment to provide warning information to operators with high risk levels in the area.

[0040] Further, as a preferred specific implementation of this embodiment, a wind direction warning method based on a CART decision tree model is proposed, comprising the following steps: S101. Collecting a data set including heart rate HR, blood pressure BP, skin temperature ST, and fatigue index FI, and marking a risk level for each sample, wherein the risk level is at least low risk, medium risk, and high risk; S102. According to the Gini impurity formula: Calculate Gini impurity; S103. Traverse all features and possible segmentation points and calculate information gain: , wherein the represents the subset after feature A is segmented, m represents the number of segmented subsets, Indicates The number of your own samples, Representation dataset The total number of samples in ; S104. Perform feature importance scoring: , wherein the represents all nodes of feature A, represents the number of samples of node t, Represents the total number of samples.

[0041] Preferably, in order to prevent overfitting, the tree needs to be pruned, and step S105 is added: , wherein the Representation Tree The error rate, Represents the cost complexity, Indicates the number of nodes in the tree. represents the complexity parameter, Represents a tree.

[0042] Exemplarily, the information gain calculation can be replaced by: , the above and Respectively represent sub-datasets with high and low heart rates. By calculating and comparing the information gain, it is determined whether the data set should be segmented according to the heart rate. The same method can be applied to other features. For blood pressure, skin temperature, and fatigue index, their Gini impurity and information gain can be calculated similarly to decide which feature is more suitable for segmenting the data set.

[0043] Furthermore, as a preferred specific implementation of this embodiment, a historical risk warning method is proposed, comprising the following steps: S201. Count the total number of times that power distribution devices are stolen in the area, calculate the frequency of power distribution devices being stolen during a specified period of time, and evaluate the degree of wear and tear of the power distribution devices according to the length of time that the power distribution devices have not been repaired / updated; S202. Calculate the risk score in each area. The specific calculation method is as follows: ; S203. Based on the risk score, the area is divided into different risk levels, for example: Low risk: score ≤ X; Medium risk: X<score ≤ Y; High risk: score>Y; S204. Send a warning signal.

[0044] Further, for the above implementation, an exemplary solution is provided: The data for a specific area is as follows: Total number of historical electricity thefts: 10 times; Historical electricity theft density: 0.5 times / month; Danger level of distributor: 2 years (no repair / update); The weight is set to: total number of electricity thefts , power theft density , distribution equipment hazard level , calculate the risk score: risk score = 0.3*10+0.4*0.5+0.3*2=3.2; according to the set risk level standard, this area can be divided into the corresponding risk level and take corresponding safety measures.

[0045] This application is explained from the perspectives of purpose of use, effectiveness, progress and novelty. The practical progress it has is in line with the functional enhancement and usage requirements emphasized by the Patent Law. The above description and drawings of this application are only the preferred embodiments of this application, and are not intended to limit this application. Therefore, all structures, devices, features, etc. that are similar or identical to those of this application, that is, all equivalent replacements or modifications made in accordance with the scope of the patent application of this application, should fall within the scope of protection of the patent application of this application.

[0046] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. An intelligent safety management and early warning system for electric power safety tools, comprising a biometric collection device and a safety early warning terminal, wherein the biometric collection device is integrated with a sensor group for collecting biological signs and positioning, and the safety early warning terminal is connected to the biometric collection device through cloud collaboration, characterized in that: The safety warning terminal is integrated with a risk warning subsystem, which is used to perform a sign risk assessment based on an artificial intelligence algorithm model on the biological sign data collected by the sensor group, establish an assessment level, and make a comprehensive judgment and warning on the sign risk of the operator according to the assessment results obtained by the algorithm model; The safety warning terminal is integrated with a risk prediction subsystem, which is used to obtain the historical danger characteristics of the positioning area according to the positioning information collected by the sensor group, determine the risk level of the area according to the number of historical danger characteristics, and send a warning signal in combination with the regional risk level; The risk prediction subsystem specifically includes: A historical area model grade assessment establishment module is used to classify the risk grade of each area in the historical area model according to the historical hazard characteristics; A positioning data acquisition module is used to acquire the positioning data collected by the sensor group, and acquire the corresponding area of ​​the positioning area and the historical area model according to the positioning data; The regional risk level warning module is used to issue warnings based on the risk level of the area where the positioning data is located; The historical regional model grade evaluation establishment module specifically includes: The risk scoring unit is used to perform risk scoring on the total number of electricity thefts in the region, the density of electricity theft in the region, and the danger level of the historical electricity theft in the region through sensitivity analysis; A risk level classification unit, used to classify the area into different risk levels based on the score, wherein the risk levels are low risk area, medium risk area and high risk area; Wherein, the risk scoring unit includes: The risk scoring model establishment subunit is used to establish a risk scoring model based on the total number of electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region; A Monte Carlo simulation subunit, used for generating parameter combination samples according to Monte Carlo simulation; The risk scoring subunit is used to calculate the corresponding risk score for the generated parameter combination sample according to the risk scoring model.

2. The intelligent safety management and early warning system for electric power safety tools according to claim 1 is characterized in that: The risk early warning subsystem specifically includes: A biological sign data acquisition module is used to acquire the biological sign data collected by the sensor group and normalize the data; A feature extraction module is used to extract features from the biometric data after data processing, and the extracted features include at least heart rate, blood pressure, body temperature, and fatigue index; A risk level establishment module, used to assign a risk level to the extracted features, wherein the risk level is at least one of low risk, medium risk, and high risk; The decision tree building module is used to traverse all features and segmentation points through the CART algorithm, and to perform data segmentation according to the best segmentation method of the Gini impurity criterion, and to recursively segment the subsets generated by each segmentation until the preset conditions are met.

3. The intelligent safety management and early warning system for electric power safety tools according to claim 1 is characterized in that: In the risk warning subsystem, the physical sign risk assessment based on the artificial intelligence algorithm model is specifically performed by calculating the information gain to segment the data set, namely: ； ； ； ； Among them, the , , and Respectively represent heart rate, blood pressure, body temperature and fatigue index, which are obtained through biological sign data. and Respectively represent the sub-datasets of high heart rate and low heart rate, the and Respectively represent the sub-datasets of high blood pressure and low blood pressure, and Respectively represent the sub-datasets of high body temperature and low body temperature, the and represent the sub-datasets with high fatigue index and low fatigue index respectively, m represents the number of divided subsets, Indicates The number of samples in the subset, Representation dataset The total number of samples in represents the Gini impurity.

4. The intelligent safety management and early warning system for electric power safety tools as claimed in claim 2, characterized in that: The specific process of data segmentation in the decision tree construction module is as follows: Evaluate the split point, traverse all features and split points, and select the split method that minimizes the Gini impurity, where Representation dataset Middle The proportion of classes Represents a dataset composed of features.

5. The intelligent safety management and early warning system for electric safety tools according to any one of claims 1 or 4, characterized in that: The hazard characteristics at least include the total number of historical electricity thefts in the area, the historical density of electricity theft in the area, and the historical hazard level of distributors in the area; the total number of historical electricity thefts in the area indicates the total number of times the distributors in the area have been stolen, the historical density of electricity theft in the area indicates the frequency of theft of electricity to the distributors in the area within a specified time period, the historical hazard level of distributors in the area indicates the degree of strain on the distributors in the area, and the historical hazard level of distributors in the area is determined by the length of time the distributors have not been repaired / updated.

6. The intelligent safety management and early warning system for electric power safety tools according to claim 1, characterized in that: The risk level classification unit includes: The variance Gini coefficient sub-unit is used to calculate the variance Gini coefficient corresponding to the total number of electricity thefts in the region, the density of electricity theft in the region, and the danger level of the historical electricity theft in the region; A threshold setting subunit is used to set the thresholds corresponding to the total number of historical electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region according to the calculated variance Gini coefficient; The risk level classification subunit is used to classify the risk level according to the set threshold value combined with the risk score calculated by the risk scoring subunit.

7. The intelligent safety management and early warning system for electric power safety tools as claimed in claim 6, characterized in that: In the variance Gini coefficient subunit, the calculation process of the variance Gini coefficient corresponding to the total number of historical electricity thefts in the region, the historical electricity theft density in the region, and the historical danger level of the distributor in the region is as follows: ； ； Among them, the represents the risk score, represents the total variance of the risk score, Represents conditional expectation The variance of When the value of The degree of change of the mean value, M represents the parameter The total number of values, j represents the total number of historical power thefts in the area The index of the value of Representation parameters The jth specific value of Representation parameters equal When the risk score The expected value of Represents risk score The average value, They respectively represent the total number of historical electricity thefts in the region or the historical density of electricity theft in the region or the historical danger level of the distribution device in the region.

8. The intelligent safety management and early warning system for electric power safety tools according to claim 1, characterized in that: The warning of the risk warning subsystem is: the risk warning subsystem uses the biological sign collection equipment to issue warning information to operators with high biological sign risks.

9. The intelligent safety management and early warning system for electric power safety tools according to claim 1, characterized in that: The early warning of the risk prediction subsystem is as follows: the risk early warning subsystem uses biological sign collection equipment to provide early warning information to operators with high risk levels in the area.

Images