Power enterprise key operation area personnel safety monitoring method and system

Abstract

The invention discloses a power enterprise key operation area personnel safety monitoring method and system, and belongs to the technical field of personnel safety monitoring, and the method comprises the steps: obtaining the geographic position information and business characteristics of a power enterprise key operation area, employing a spatial clustering algorithm to carry out the adaptive division of the operation area, and obtaining a plurality of management units; during system architecture design, a distributed computing and storage framework is adopted, position data and state information are subjected to fragmentation storage and parallel processing according to the management unit, the concurrent processing capacity and response speed of the system are improved, and the performance requirement of large-scale personnel and equipment monitoring is met; an abnormal detection model of the personnel equipment state is established, abnormal behavior identification is performed by analyzing historical data, summarizing a normal mode of the personnel equipment state and utilizing a machine learning algorithm, and once an abnormal condition is found, early warning is performed in time and emergency treatment measures are taken, so that the safety and the reliability of the system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of personnel safety monitoring, and in particular relates to a method and system for monitoring personnel safety in key operating areas of an electric power enterprise. Background Art

[0002] The personnel safety monitoring system in the key operation areas of power enterprises faces the contradiction between regional division and personnel and equipment management when managing different operation areas in a unified manner. On the one hand, in order to achieve refined management, it is necessary to divide the operation area in detail, accurately locate and monitor the personnel and equipment in the area in real time. On the other hand, too detailed regional division will increase the complexity of management. The system needs to process a large amount of location data and status information, which puts higher requirements on the performance and reliability of the system. At the same time, there is often interaction and flow between personnel and equipment in different areas. How to ensure clear regional division while being able to flexibly handle cross-regional personnel and equipment management is a technical problem that needs to be solved urgently. In addition, in the actual operation environment, the location and status of personnel and equipment in the area may change at any time. How to achieve real-time dynamic regional division and management is also a challenging problem. The system needs to ensure the consistency and continuity of regional management strategies while meeting the requirements of real-time and accuracy to avoid management blind spots and conflicts. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes a method and system for monitoring personnel safety in key operation areas of an electric power enterprise, so as to solve the above problems existing in the prior art.

[0004] To achieve the above object, the present invention provides a method for monitoring personnel safety in key operation areas of a power enterprise, comprising:

[0005] Obtaining geographic location coordinate data and business characteristic description information of the key operation area of ​​the power enterprise, dividing the key operation area of ​​the power enterprise based on the geographic location coordinate data and business characteristic description information, and generating a plurality of management units;

[0006] Deploy sensors inside the plurality of management units, collect location information of personnel and operation information of equipment based on the sensors, pre-process the location information and the operation information to obtain a pre-processed data set;

[0007] Constructing a region switching detection algorithm, detecting the location information of the personnel based on the region switching detection algorithm, and generating personnel region switching monitoring results;

[0008] Constructing an anomaly detection model for the status of personnel and equipment, identifying the preprocessed data set based on the anomaly detection model, and generating an abnormal situation monitoring result;

[0009] Based on the personnel area switching monitoring results and the abnormal situation monitoring results, personnel safety is identified and warned.

[0010] Preferably, the process of generating a plurality of management units includes:

[0011] Obtain geographic location coordinate data and business characteristics description information of key operating areas of power companies;

[0012] Acquire a spatial distance matrix between operation areas based on the geographic location coordinate data;

[0013] Constructing a business characteristic vector of the operation area based on the business characteristic description information;

[0014] The spatial distance matrix and the business characteristic vector are calculated based on a spatial clustering algorithm, and the operation area is adaptively divided based on the calculation result to obtain the plurality of management units.

[0015] Preferably, the process of obtaining the preprocessed data set includes:

[0016] Deploy positioning devices and sensor units on key areas and key equipment of the several management units, and obtain the position coordinates and working status of personnel and equipment based on the positioning devices and the sensor units;

[0017] Preprocessing the position coordinates and working status of the personnel and equipment to obtain a data set to be fused;

[0018] The data set to be fused is processed based on a data fusion algorithm to obtain the preprocessed data set. Preferably, the location coordinates and working status of the personnel and equipment are preprocessed to obtain the data set to be fused, including:

[0019] Performing denoising processing on the position coordinates based on a Kalman filter algorithm to generate denoised position coordinates;

[0020] Normalizing the working state based on the Min-Max normalization algorithm to obtain normalized data;

[0021] Based on the DS evidence theory algorithm, the denoised position coordinates and the normalized data are subjected to correlation analysis to obtain a correlation analysis result;

[0022] The denoised position coordinates and the normalized data are integrated based on the association analysis result to obtain the data set to be fused.

[0023] Preferably, the process of generating personnel area switching monitoring results includes:

[0024] Generate location change trajectory data of the personnel based on the location information of the personnel;

[0025] Determining the position change trajectory data based on the area division information of the plurality of management units to obtain a change determination result;

[0026] Analyzing the change determination result based on a machine learning algorithm for trajectory shape recognition to generate a region switching pattern;

[0027] The length of stay of the personnel is determined based on the time series characteristics of the personnel trajectory and the time series analysis algorithm, and the personnel area switching monitoring result is generated based on the length of stay of the personnel and the area switching mode.

[0028] Preferably, the process of constructing an anomaly detection model of personnel and equipment status includes:

[0029] Acquire historical data of personnel and equipment status, and perform cleaning and standardization preprocessing on the historical data of personnel and equipment status to obtain an equipment status data set;

[0030] A long short-term memory network is constructed, and the long short-term memory network is trained using the device status data set to obtain the anomaly detection model.

[0031] Preferably, the process of identifying the preprocessed data set based on the anomaly detection model and generating an abnormal situation monitoring result includes:

[0032] Inputting the preprocessed data set into the anomaly detection model for calculation to obtain a calculation result;

[0033] The calculation result is judged based on the isolation forest algorithm, and the abnormal situation monitoring result is generated based on the judgment result.

[0034] To achieve the above objectives, the present invention also provides a personnel safety monitoring system for key operation areas of a power enterprise, comprising:

[0035] A division module, used to obtain geographic location coordinate data and business characteristic description information of the key operation area of ​​the power enterprise, divide the key operation area of ​​the power enterprise based on the geographic location coordinate data and business characteristic description information, and generate a plurality of management units;

[0036] A data acquisition module, configured to deploy sensors inside the plurality of management units, collect personnel location information and equipment operation information based on the sensors, pre-process the location information and the operation information, and obtain a pre-processed data set;

[0037] A location detection module is used to construct an area switching detection algorithm, detect the location information of the personnel based on the area switching detection algorithm, and generate a personnel area switching monitoring result;

[0038] A situation detection module, used to construct an abnormality detection model of personnel and equipment status, identify the preprocessed data set based on the abnormality detection model, and generate abnormal situation monitoring results;

[0039] The identification and warning module is used to identify and warn personnel safety based on the personnel area switching monitoring results and the abnormal situation monitoring results.

[0040] Compared with the prior art, the present invention has the following advantages and technical effects:

[0041] The present invention discloses an adaptive management method for the operation area of ​​an electric power enterprise. The method first obtains the geographical location and business characteristics of the key operation area, and uses a spatial clustering algorithm to perform adaptive division to form a management unit. High-precision positioning equipment and sensors are deployed in the unit to collect personnel and equipment information in real time, and a unified status information table is generated through data fusion. In view of the problem of cross-regional mobility, a switching detection algorithm is designed to trigger management strategy adjustment. An incremental learning mechanism is introduced to dynamically adjust unit division and update status information. A distributed architecture is used to improve concurrent processing capabilities. An anomaly detection model is constructed to timely warn and deal with abnormal situations. The present invention realizes intelligent management of the operation area, improves the accuracy and efficiency of personnel and equipment monitoring, enhances the safety and reliability of the system, and provides strong support for the refined management of electric power enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0043] Figure 1 This is a flow chart of a method for monitoring personnel safety in key operating areas of a power enterprise according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0045] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] Embodiment 1

[0047] like Figure 1 As shown, this embodiment provides a method for monitoring personnel safety in key operation areas of a power enterprise, including:

[0048] Obtaining geographic location coordinate data and business characteristic description information of the key operation area of ​​the power enterprise, dividing the key operation area of ​​the power enterprise based on the geographic location coordinate data and business characteristic description information, and generating a plurality of management units;

[0049] Deploy sensors inside the plurality of management units, collect location information of personnel and operation information of equipment based on the sensors, pre-process the location information and the operation information to obtain a pre-processed data set;

[0050] Constructing a region switching detection algorithm, detecting the location information of the personnel based on the region switching detection algorithm, and generating personnel region switching monitoring results;

[0051] Constructing an anomaly detection model for the status of personnel and equipment, identifying the preprocessed data set based on the anomaly detection model, and generating an abnormal situation monitoring result;

[0052] Based on the personnel area switching monitoring results and the abnormal situation monitoring results, personnel safety is identified and warned.

[0053] Specifically:

[0054] Step S101, obtaining geographical location information and business characteristics of key operating areas of the power enterprise, and using a spatial clustering algorithm to adaptively divide the operating areas to obtain a number of management units.

[0055] Step S102, deploy high-precision positioning equipment and sensors inside the management unit to collect the location coordinates, working status and other information of personnel and equipment in real time, process the collected multi-source heterogeneous data through data fusion algorithm, and generate a unified personnel and equipment status information table as a basis for subsequent management and decision-making.

[0056] Step S103, in order to solve the problem of cross-regional movement of personnel and equipment between the management units, a regional switching detection algorithm is designed to determine whether a regional switching has occurred by analyzing the position change trajectory of the personnel and equipment. If a regional switching occurs, the corresponding management strategy adjustment is triggered to ensure the continuity and consistency of management.

[0057] Step S104, in order to adapt to the dynamic changes of the working environment, an incremental learning mechanism is introduced to dynamically adjust the division of the management unit according to the newly collected personnel and equipment location and status data, and at the same time update the personnel and equipment status information table inside the management unit to ensure the matching degree between the management strategy and the actual situation.

[0058] Step S105, when designing the system architecture, a distributed computing and storage framework is adopted to perform shard storage and parallel processing of the location data and status information according to the management units, so as to improve the concurrent processing capability and response speed of the system and meet the performance requirements of large-scale personnel and equipment monitoring.

[0059] Step S106, build an abnormal detection model for personnel and equipment status, analyze historical data, summarize the normal patterns of personnel and equipment status, use machine learning algorithms to identify abnormal behaviors, and once an abnormal situation is found, timely warn and take emergency measures to improve the safety and reliability of the system.

[0060] Furthermore, the process of generating a plurality of management units includes:

[0061] Obtain geographic location coordinate data and business characteristics description information of key operating areas of power companies;

[0062] Acquire a spatial distance matrix between operation areas based on the geographic location coordinate data;

[0063] Constructing a business characteristic vector of the operation area based on the business characteristic description information;

[0064] The spatial distance matrix and the business characteristic vector are calculated based on a spatial clustering algorithm, and the operation area is adaptively divided based on the calculation result to obtain the plurality of management units.

[0065] Specifically:

[0066] Obtain the geographic location coordinate data and business characteristics description information of the key operating areas of the power enterprise, and build a geographic spatial database of the operating area. According to the geographic location coordinate data of the operating area, calculate the spatial distance matrix between the operating areas. According to the business characteristics description information of the operating area, use text mining technology to extract keywords and build the business characteristics vector of the operating area. Use the spatial distance matrix and business characteristics vector of the operating area as the input of the spatial clustering algorithm, and use the hierarchical clustering algorithm to adaptively divide the operating area. By setting the cluster number threshold of the clustering algorithm, the number of generated management units is controlled to achieve adaptive division of the operating area. For each management unit obtained by division, calculate its geographic center point coordinates as the geographic location attribute of the management unit. For each management unit obtained by division, count the business characteristics keywords of the operating area it contains as the business characteristics attribute of the management unit.

[0067] Specifically, in order to obtain the geographic location coordinate data and business characteristics description information of the key operation areas of power enterprises, the global positioning system (GPS) can be used to locate the operation area and obtain its latitude and longitude coordinates. At the same time, through on-site surveys and interviews with business personnel, the business characteristics information of the operation area, such as operation type, operation intensity, operation frequency, etc., is collected. These data are entered into the geographic information system (GIS) to build a geographic spatial database of the operation area. The spatial analysis function of GIS is used to calculate the Euclidean distance between the operation areas and generate a spatial distance matrix. Text mining is performed on the business characteristics description information of the operation area, and the TF-IDF algorithm is used to extract keywords to generate a business characteristics vector. The vector dimension is the number of keywords, and each element represents the weight of the keyword. The spatial distance matrix and the business characteristics vector are input into the hierarchical clustering algorithm, and the Ward method is used to calculate the distance between clusters. By setting the cluster number threshold to 10, the number of generated management units is controlled to achieve adaptive division of the operation area. For each management unit, the arithmetic mean of the geographic coordinates of the operation area it contains is calculated as the coordinates of the geographic center point of the management unit. For each management unit, the business characteristic keywords of the operating areas it contains are counted, sorted from high to low according to the word frequency, and the top 5 keywords are selected as the business characteristic attributes of the management unit.

[0068] Furthermore, the process of obtaining the preprocessed data set includes:

[0069] Deploy positioning devices and sensor units on key areas and key equipment of the several management units, and obtain the position coordinates and working status of personnel and equipment based on the positioning devices and the sensor units;

[0070] Preprocessing the position coordinates and working status of the personnel and equipment to obtain a data set to be fused;

[0071] The data set to be fused is processed based on a data fusion algorithm to obtain the preprocessed data set.

[0072] Specifically:

[0073] High-precision positioning equipment and various types of sensors are installed and deployed in key areas and key equipment within the management unit to collect and monitor the location coordinates, working status and operating parameters of personnel in real time. The collected multi-source heterogeneous data such as personnel location coordinate data, working status information, equipment operating parameters, etc. are transmitted to the data processing center. In the data processing center, data cleaning and data preprocessing technologies are used to perform denoising and normalization on the collected raw data to improve data quality. According to the pre-established data fusion rules and algorithm models, the processed multi-source heterogeneous data are subjected to correlation analysis and fusion calculation to generate a unified personnel and equipment status information table.

[0074] Furthermore, the process of preprocessing the position coordinates and working status of the personnel and equipment to obtain the data set to be fused includes:

[0075] Performing denoising processing on the position coordinates based on a Kalman filter algorithm to generate denoised position coordinates;

[0076] Normalizing the working state based on the Min-Max normalization algorithm to obtain normalized data;

[0077] Based on the DS evidence theory algorithm, the denoised position coordinates and the normalized data are subjected to correlation analysis to obtain a correlation analysis result;

[0078] The denoised position coordinates and the normalized data are integrated based on the association analysis result to obtain the data set to be fused.

[0079] Specifically, high-precision GPS positioning equipment and various types of sensors, such as temperature sensors, humidity sensors, and vibration sensors, are installed and deployed in key areas and key equipment within the management unit. The GPS positioning equipment collects the position coordinates of personnel in real time at a frequency of 1 second, with an accuracy of up to centimeters. The sensor monitors the operating parameters of the equipment, such as temperature, humidity, and vibration, in real time at a frequency of 100 milliseconds. The collected data is transmitted to the data processing center at a rate of 10 megabits per second through a wireless transmission module. In the data processing center, the Kalman filter algorithm is used to denoise the GPS position coordinate data, remove abnormal points, and improve positioning accuracy. The Min-Max normalization algorithm is used to normalize the equipment operating parameter data collected by the sensor, and the data of different dimensions are unified to between 0 and 1 for subsequent analysis. Then, according to the pre-established data fusion rules, the DS evidence theory algorithm is used to perform correlation analysis and fusion calculation on the processed personnel position coordinate data, work status information, and equipment operating parameters to generate a unified personnel and equipment status information table.

[0080] According to the preset data fusion rules, the processed multi-source heterogeneous data is obtained, and the classification algorithm is used to classify the data according to the characteristic attributes of the data to obtain the classified data set; according to the classified data set, the association rule mining algorithm is used to mine the association relationship between the data to obtain the association relationship rule base; according to the association relationship rule base, the decision tree algorithm is used to perform association analysis on the data to obtain the association analysis results; according to the association analysis results, the preset data fusion algorithm model is obtained, and the fused data is obtained through fusion calculation; according to the fused data, the preset unified format template is obtained, and the data is mapped to the template to obtain the data in a unified format; according to the data in the unified format, the personnel status and equipment status information are extracted to obtain the personnel status information and the equipment status information; according to the personnel status information and the equipment status information, the personnel equipment status information table is generated to complete the fusion processing of multi-source heterogeneous data.

[0081] The Apriori association rule mining algorithm is used to mine the association rules between abnormal behaviors and potential faults with a minimum support of 05 and a minimum confidence of 8. It is found that the probability of abnormal behaviors A and B occurring simultaneously is 75%, and the probability of abnormal behaviors C and potential faults D occurring simultaneously is 65%. Finally, based on the mined association rules, early warning rules are generated, such as: when personnel and equipment trigger abnormal behaviors A and B at the same time, a red warning is triggered; when equipment triggers abnormal behavior C and triggers potential fault D within 30 minutes, an orange warning is triggered, thus achieving real-time intelligent monitoring and early warning of abnormal behaviors and potential faults of personnel and equipment.

[0082] Furthermore, the process of generating the personnel area switching monitoring result includes:

[0083] Generate location change trajectory data of the personnel based on the location information of the personnel;

[0084] Determining the position change trajectory data based on the area division information of the plurality of management units to obtain a change determination result;

[0085] Analyzing the change determination result based on a machine learning algorithm for trajectory shape recognition to generate a region switching pattern;

[0086] The length of stay of the personnel is determined based on the time series characteristics of the personnel trajectory and the time series analysis algorithm, and the personnel area switching monitoring result is generated based on the length of stay of the personnel and the area switching mode.

[0087] Specifically, by deploying high-precision positioning devices such as UWB positioning systems in the management unit, the location coordinates of personnel and equipment can be obtained in real time to form continuous trajectory data. The regional division information of the management unit is pre-entered into the system. By comparing the starting and ending coordinates of the device trajectory with the boundary coordinates of the management unit, it can be determined whether cross-region movement occurs. If a device moves from area A to area B, and its trajectory starts in area A and ends in area B, the regional switching detection process is triggered. The trajectory shape recognition model based on the DTW (dynamic time warping) algorithm is used to extract and classify the shape features of the device motion trajectory, match it with the preset regional switching pattern library, and identify the trajectory that meets the switching pattern. At the same time, the sliding window algorithm is used to perform time series analysis on the trajectory data, and the length of time the device stays in each management unit is counted. If the length of time exceeds 30 minutes, the validity of the regional switching behavior is further verified. According to the regional switching detection results, the system automatically matches the management policy corresponding to the target management unit from the policy library. For example, after the device in area A enters area B, the device control rules and security policies in area B are automatically loaded to achieve adaptive switching of management policies. During the entire process of cross-regional movement of equipment, the system triggers regional judgment and policy matching every 5 seconds to ensure real-time and continuous management.

[0088] Furthermore, the process of constructing the anomaly detection model of personnel and equipment status includes:

[0089] Acquire historical data of personnel and equipment status, and perform cleaning and standardization preprocessing on the historical data of personnel and equipment status to obtain an equipment status data set;

[0090] A long short-term memory network is constructed, and the long short-term memory network is trained using the device status data set to obtain the anomaly detection model.

[0091] Clean the collected data, remove invalid or erroneous data records, process missing values, remove noise, etc.

[0092] Data standardization is performed so that data of different dimensions and ranges can be compared under the same standard, which usually includes normalization or standardization. Normalization is to scale the data to the interval [0,1], while standardization is to convert the data into a distribution with a mean of 0 and a standard deviation of 1. When standardizing, the statistical parameters (mean and standard deviation) of the training set should be used to transform the test set data to prevent data leakage and ensure the effectiveness of model evaluation. Design the LSTM network structure, including determining the number and size of hidden layers, selecting appropriate activation functions, etc. LSTM models usually include an input layer, one or more LSTM layers, and an output layer. The LSTM layer controls the flow of information through its gating mechanism (forget gate, input gate, output gate) to capture long-term dependencies in time series data. Use the processed data set to train the LSTM model. During the training process, the model learns to recognize the normal pattern of device status. The mean square error loss function is adopted, and the Adam optimizer is used to train the model.

[0093] Furthermore, the process of identifying the preprocessed data set based on the anomaly detection model and generating an abnormal situation monitoring result includes:

[0094] Inputting the preprocessed data set into the anomaly detection model for calculation to obtain a calculation result;

[0095] The calculation result is judged based on the isolation forest algorithm, and the abnormal situation monitoring result is generated based on the judgment result.

[0096] When an abnormal equipment status is detected, the system automatically generates warning information, which is divided into three levels according to the abnormality score: mild (6-7), moderate (7-8), and severe (8 or above), and sent to the on-duty personnel through SMS, email, etc. At the same time, the corresponding emergency plan is activated according to the abnormality level. For example, in the case of severe abnormality, the equipment is automatically switched to the backup system, and the on-site engineer is notified for maintenance. The detailed information of the abnormal event will be recorded in the case library, including the time of abnormal occurrence, abnormal characteristics, disposal measures, recovery time, etc., for subsequent model optimization and plan improvement. Every month, the system will automatically retrain the anomaly detection model with the latest historical data, and use the cross-validation method to evaluate the model performance, and continuously improve the accuracy and recall rate of anomaly recognition. Through the above measures, the equipment status monitoring system can detect abnormal conditions of the equipment in real time and take timely countermeasures to minimize the impact of abnormalities on production operations and improve the safety and reliability of the system.

[0097] Embodiment 2

[0098] This embodiment provides a personnel safety monitoring system for key operation areas of a power enterprise, including:

[0099] A division module, used to obtain the geographical location coordinate data and business characteristic description information of the key operation area of ​​the power enterprise, divide the key operation area of ​​the power enterprise based on the geographical location coordinate data and business characteristic description information, and generate a plurality of management units;

[0100] A data acquisition module, configured to deploy sensors inside the plurality of management units, collect location information of personnel and operation information of equipment based on the sensors, pre-process the location information and the operation information, and obtain a pre-processed data set;

[0101] A location detection module is used to construct an area switching detection algorithm, detect the location information of the personnel based on the area switching detection algorithm, and generate a personnel area switching monitoring result;

[0102] A situation detection module, used to construct an abnormality detection model of personnel and equipment status, identify the preprocessed data set based on the abnormality detection model, and generate abnormal situation monitoring results;

[0103] The identification and warning module is used to identify and warn personnel safety based on the personnel area switching monitoring results and the abnormal situation monitoring results.

[0104] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for monitoring personnel safety in key operation areas of a power enterprise, characterized in that: The following steps are involved: Obtaining geographic location coordinate data and business characteristic description information of the key operation area of ​​the power enterprise, dividing the key operation area of ​​the power enterprise based on the geographic location coordinate data and business characteristic description information, and generating a plurality of management units; Deploy sensors inside the plurality of management units, collect location information of personnel and operation information of equipment based on the sensors, pre-process the location information and the operation information to obtain a pre-processed data set; Constructing a region switching detection algorithm, detecting the location information of the personnel based on the region switching detection algorithm, and generating personnel region switching monitoring results; Constructing an anomaly detection model for the status of personnel and equipment, identifying the preprocessed data set based on the anomaly detection model, and generating an abnormal situation monitoring result; Based on the personnel area switching monitoring results and the abnormal situation monitoring results, personnel safety is identified and warned.

2. The method for monitoring personnel safety in key operation areas of a power enterprise according to claim 1, characterized in that: The process of generating a plurality of management units includes: Obtain geographic location coordinate data and business characteristics description information of key operating areas of power companies; Acquire a spatial distance matrix between operation areas based on the geographic location coordinate data; Constructing a business characteristic vector of the operation area based on the business characteristic description information; The spatial distance matrix and the business characteristic vector are calculated based on a spatial clustering algorithm, and the operation area is adaptively divided based on the calculation result to obtain the plurality of management units.

3. The method for monitoring personnel safety in key operation areas of a power enterprise according to claim 1, characterized in that: The process of obtaining the preprocessed data set includes: Deploy positioning devices and sensor units on key areas and key equipment of the several management units, and obtain the position coordinates and working status of personnel and equipment based on the positioning devices and the sensor units; Preprocessing the position coordinates and working status of the personnel and equipment to obtain a data set to be fused; The data set to be fused is processed based on a data fusion algorithm to obtain the preprocessed data set.

4. The method for monitoring personnel safety in key operation areas of a power enterprise according to claim 3 is characterized in that: The process of preprocessing the position coordinates and working status of the personnel and equipment to obtain the data set to be fused includes: Performing denoising processing on the position coordinates based on a Kalman filter algorithm to generate denoised position coordinates; Normalizing the working state based on the Min-Max normalization algorithm to obtain normalized data; Based on the DS evidence theory algorithm, the denoised position coordinates and the normalized data are subjected to correlation analysis to obtain a correlation analysis result; The denoised position coordinates and the normalized data are integrated based on the association analysis result to obtain the data set to be fused.

5. The method for monitoring personnel safety in key operation areas of a power enterprise according to claim 1, characterized in that: The process of generating the personnel area switching monitoring result includes: Generate location change trajectory data of the personnel based on the location information of the personnel; Determining the position change trajectory data based on the area division information of the plurality of management units to obtain a change determination result; Analyzing the change determination result based on a machine learning algorithm for trajectory shape recognition to generate a region switching pattern; The length of stay of the personnel is determined based on the time series characteristics of the personnel trajectory and the time series analysis algorithm, and the personnel area switching monitoring result is generated based on the length of stay of the personnel and the area switching mode.

6. The method for monitoring personnel safety in key operation areas of a power enterprise according to claim 1, characterized in that: The process of constructing the anomaly detection model of personnel and equipment status includes: Acquire historical data of personnel and equipment status, and perform cleaning and standardization preprocessing on the historical data of personnel and equipment status to obtain an equipment status data set; A long short-term memory network is constructed, and the long short-term memory network is trained using the device status data set to obtain the anomaly detection model.

7. The method for monitoring personnel safety in key operation areas of a power enterprise according to claim 1, characterized in that: The process of identifying the preprocessed data set based on the anomaly detection model and generating an abnormal situation monitoring result includes: Inputting the preprocessed data set into the anomaly detection model for calculation to obtain a calculation result; The calculation result is judged based on the isolation forest algorithm, and the abnormal situation monitoring result is generated based on the judgment result.

8. The monitoring system of the method for monitoring personnel safety in key operation areas of a power enterprise according to any one of claims 1 to 7, characterized in that: include: A division module, used to obtain geographic location coordinate data and business characteristic description information of the key operation area of ​​the power enterprise, divide the key operation area of ​​the power enterprise based on the geographic location coordinate data and business characteristic description information, and generate a plurality of management units; A data acquisition module, configured to deploy sensors inside the plurality of management units, collect personnel location information and equipment operation information based on the sensors, pre-process the location information and the operation information, and obtain a pre-processed data set; A location detection module is used to construct an area switching detection algorithm, detect the location information of the personnel based on the area switching detection algorithm, and generate a personnel area switching monitoring result; A situation detection module, used to construct an abnormality detection model of personnel and equipment status, identify the preprocessed data set based on the abnormality detection model, and generate abnormal situation monitoring results; The identification and warning module is used to identify and warn personnel safety based on the personnel area switching monitoring results and the abnormal situation monitoring results.

Images