Method for positioning personnel in power enterprise factory

Abstract

The invention discloses a personnel positioning method in a power enterprise factory, and the method comprises the steps: obtaining the position coordinate data of an employee, and carrying out the preprocessing of the position coordinate data of the employee; integrating the preprocessed coordinate data to obtain working group coordinate data, and obtaining a working group working area according to the working group coordinate data; obtaining timestamps of the position coordinate data of the personnel, obtaining stay time distribution according to the timestamps of the position coordinate data of the personnel, and dividing and optimizing overlapped areas of the working areas of the working groups according to the stay time distribution to obtain mergeable areas; and acquiring the real-time position of the staff position, and judging the real-time position according to the independent area and the mergeable area of the working area of the working group to obtain a real-time position judgment result so as to judge whether the staff is in the working area or not.

Description

Technical Field

[0001] The invention belongs to the technical field of personnel positioning and identification, and in particular relates to a personnel positioning method within a power enterprise plant. Background Art

[0002] In the plant management of power enterprises, it is necessary to determine the relevant positions of employees within the enterprise to determine whether they are within the working range. The determination of personnel in the working area of ​​the power will have an impact on the management of the enterprise and the safety of personnel. However, after receiving the employee position coordinate data, it is necessary to compare it with the preset employee work position coordinate range. The technical difficulty here is that the work position coordinate ranges of different employees may overlap. However, the simple area division method is not accurate for the positioning of personnel. In addition, if the actual work position of an employee exceeds the preset coordinate range, but overlaps with the range of other employees, it will also lead to misjudgment. How the system can reasonably divide the work coordinate range of each employee and accurately determine whether the employee's position is in the corresponding area to ensure the effectiveness and safety of the plant management of power enterprises is a technical problem to be solved. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes a method for locating personnel within a power enterprise plant to solve the above problems existing in the prior art.

[0004] To achieve the above object, the present invention provides a method for locating personnel in a power enterprise plant, comprising:

[0005] Obtaining employee location coordinate data and preprocessing the employee location coordinate data;

[0006] Integrate the preprocessed coordinate data to obtain the work group coordinate data, and obtain the work group working area according to the work group coordinate data;

[0007] Obtain and obtain the residence time distribution according to the timestamp of the personnel's location coordinate data, and divide and optimize the overlapping areas of the work group's work area according to the residence time distribution to obtain the mergeable areas;

[0008] The real-time location of the employee is obtained, and the real-time location is judged according to the independent area and the mergeable area of ​​the work group's work area to obtain the real-time location judgment result to realize the judgment result of whether the person is in the work area.

[0009] Optionally, the process of preprocessing the employee location coordinate data includes:

[0010] The employee location coordinate data is converted to a standard coordinate system, filtered and clustered. The filtering includes: performing threshold detection on the unified coordinate data through an outlier detection method, and removing the coordinate data that does not meet the threshold; the clustering analysis includes: clustering the filtered coordinate data through a DBSCAN density clustering algorithm to obtain the coordinate data of each employee corresponding to the coordinate cluster, and performing threshold detection on the coordinate data in the coordinate cluster again to obtain the preprocessed coordinate data.

[0011] Optionally, the process of obtaining the work area of ​​the work group includes:

[0012] The employee's job attribute information and the layout information of the workshop are obtained, and the preprocessed coordinate data are clustered into different workgroup coordinate data according to the employee's job attribute information and the layout information of the workshop by using the K-means clustering algorithm, and the workgroup coordinate data are processed by the Graham scanning algorithm to obtain a polygonal geometric model, and the polygonal geometric model is optimized according to the optimization algorithm, and the minimization of the polygonal geometric model is taken as the goal, and the overlapping area of ​​the polygonal geometric model is taken as less than or equal to a threshold as the stopping condition, so as to obtain different workgroup working areas, wherein each workgroup working area includes an overlapping area and an independent area.

[0013] Optionally, the process of obtaining the residence time distribution includes:

[0014] Based on the preprocessed coordinate data and corresponding timestamps of the overlapping area, the coordinate data and the corresponding timestamps are calculated to count the residence time distribution of different locations.

[0015] Optionally, the process of dividing the overlapping area of ​​the work group work area includes:

[0016] The similarity of the residence time distribution in different time periods and the KL divergence of the residence time distribution are calculated. According to the similarity calculation results and the KL divergence, the overlapping areas are segmented, and the segmentation results are optimized through the optimization algorithm to obtain the mergeable areas.

[0017] Optionally, the process of determining the real-time location includes:

[0018] The real-time location is judged according to the independent area. When the real-time location is in the independent area, the person is within the working range. Otherwise, the real-time location is judged according to the mergeable area. When the real-time location is in the mergeable area, the person is within the working range. Otherwise, the historical location data corresponding to the person is obtained, and cluster analysis is performed on the historical location data to obtain the regional center of gravity position. A threshold judgment is performed based on the regional center of gravity position and the real-time location. If the distance between the regional center of gravity position and the real-time location is less than the threshold, the person is within the working range. Otherwise, it is judged as not being within the working range and is an abnormal activity.

[0019] Optionally, the real-time location determination result also includes:

[0020] If the real-time position judgment result is out of the working range, an early warning message is generated, which includes the employee ID, abnormal time and location coordinates, and the trajectory is played back.

[0021] Optionally, the real-time location determination result also includes:

[0022] The decision tree algorithm is used to analyze the employees’ historical location data to determine whether there are abnormal patterns, where the abnormal patterns include situations where the time spent in the work area during non-working hours is greater than a threshold.

[0023] Compared with the prior art, the present invention has the following advantages and technical effects:

[0024] The present invention discloses a method for positioning personnel in a power enterprise plant. The method first performs standardization processing and outlier detection on the received employee position coordinate data, and then divides the standard working range according to the employee's job attributes and the workshop layout. For overlapping working ranges, subdivision is performed through historical data cluster analysis. When determining the area where the employee is located, the present invention first checks whether the coordinates are within the standard range, and if not, further determines whether they are in the overlapping area. For abnormal situations that exceed the preset range, the present invention automatically generates an early warning and starts trajectory analysis. Finally, a regional determination report containing determination results, confidence and other information is formed for applications such as attendance management. The present invention improves the accuracy and practicality of employee position determination through multi-dimensional data analysis and dynamic adjustment. At the same time, as a region with more power equipment, the determination of the position of the power enterprise can also prevent personnel from mistakenly entering areas outside the working range, thereby improving its safety to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] 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:

[0026] Figure 1 The present invention is a flowchart of a method for locating personnel within a power enterprise plant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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.

[0028] 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.

[0029] The present invention provides a method for locating personnel in a power enterprise plant, which mainly includes:

[0030] After receiving the employee's location coordinate data, it is converted into a unified coordinate system format to eliminate the coordinate deviation between different positioning devices and improve the accuracy of subsequent comparisons. At the same time, the coordinate data is detected for outliers to eliminate coordinate points that are obviously beyond the reasonable range and reduce interference;

[0031] According to the employee's job attributes and workshop layout, the standard work coordinate range of each employee is reasonably divided, and the range is described using geometric figures such as rectangles and polygons. The overlapping area between the work ranges of adjacent employees is minimized to reduce the complexity of subsequent judgments;

[0032] For overlapping work scopes of employees, the overlapping areas are further subdivided. Through cluster analysis of employees' historical location coordinates, the activity patterns and residence time distribution characteristics of each employee in the overlapping area are obtained. Based on this, the overlapping areas are divided to meet the actual work trajectory of each employee as much as possible.

[0033] When determining the area where an employee is located, first check whether the coordinate point falls within the standard working range of the employee. If so, directly determine that the employee is in his or her working area. If not, further determine whether the coordinate point is in the overlapping area.

[0034] If the coordinate point is located in the overlapping area, the overlapping area boundary obtained in step 3 is divided to determine which employee the coordinate point belongs to. At the same time, the confidence level of the attribution judgment is dynamically adjusted by referring to the current behavior status of the employee (such as moving, staying, etc.);

[0035] If the employee's position coordinates exceed all preset working ranges, they will be marked as abnormal. The system will automatically generate warning information, notify relevant responsible personnel to verify the situation in time, and start the trajectory playback and behavior analysis program to determine whether the employee's actual work trajectory has obvious abnormalities and whether it is in a dangerous area outside the working range;

[0036] Based on the above determination results, the system generates an area determination report for the employee, including the employee's basic information, determination area, determination confidence, abnormal status, etc. The report data is automatically summarized into the employee management database for subsequent attendance, performance evaluation, and safety area range determination. At the same time, the area determination results are fed back to the employee in real time to facilitate their self-management and behavior optimization.

[0037] like Figure 1 As shown, a method for locating personnel in a power enterprise plant in this embodiment may specifically include:

[0038] Step S101: After receiving the employee's location coordinate data, convert it into a unified coordinate system format to eliminate the coordinate deviation between different positioning devices and improve the accuracy of subsequent comparisons. At the same time, perform outlier detection on the coordinate data to eliminate coordinate points that are obviously beyond the reasonable range and reduce interference.

[0039] Specifically, it includes: acquiring original coordinate data collected by at least one employee positioning device, wherein the original coordinate data includes coordinate points in different coordinate system formats;

[0040] The original coordinate data is uniformly converted into standard coordinate data in a target coordinate system format by using a coordinate conversion algorithm;

[0041] Acquire the standard coordinate data, and use an outlier detection algorithm to determine whether each coordinate point in the standard coordinate data exceeds a preset normal activity area threshold, and if it exceeds the threshold, mark the coordinate point as an abnormal coordinate point;

[0042] Acquire an outlier detection result, remove coordinate points marked as abnormal in the standard coordinate data, and obtain filtered coordinate data;

[0043] Performing cluster analysis on the filtered coordinate data using a density clustering algorithm, and dividing the filtered coordinate data into at least one coordinate cluster according to a preset clustering radius and a minimum point number threshold;

[0044] For each of the at least one coordinate cluster, calculate the center coordinate point of the coordinate cluster, compare the center coordinate point with the preset employee workstation coordinates, and if the distance between the center coordinate point and the employee workstation coordinates is less than a preset threshold, associate the coordinate point data of the coordinate cluster with the employee;

[0045] Specifically, the original coordinate data collected by the employee positioning device includes coordinate points in different coordinate system formats such as WGS84, GCJ02 and BD09. The seven-parameter coordinate transformation model is used to uniformly transform the original coordinate data into standard coordinate data in the same coordinate system, such as a coordinate system based on a certain location of the enterprise, through parameters such as translation, rotation and scale transformation. After obtaining the standard coordinate data, the distance-based outlier detection algorithm is used to set the normal activity area threshold to 500 meters to determine whether each coordinate point exceeds the threshold. If it exceeds, the point is marked as an abnormal coordinate point. After removing the abnormal coordinate points, the DBSCAN density clustering algorithm is used to perform cluster analysis on the filtered coordinate data. The cluster radius is set to 10 meters and the minimum point number threshold is set to 20, and the coordinate data is divided into multiple coordinate clusters. For each coordinate cluster, its central coordinate point is calculated and compared with the preset employee workstation coordinates for Euclidean distance. If the distance is less than 5 meters, the coordinate point data of the coordinate cluster is associated with the corresponding employee.

[0046] Through the above content, we try to remove the activity locations of different employees that are not in the factory area, set the coordinates within the factory area as much as possible, and cluster the different coordinates in the factory area according to different employees to obtain the corresponding coordinate clusters of the employees in the factory area.

[0047] After being associated with the corresponding employee, the coordinates of different coordinate clusters are detected again. An outlier detection algorithm is used to determine whether each coordinate point in the standard coordinate data coordinate cluster exceeds the preset normal activity area threshold. Determine the preset normal activity area threshold; for each coordinate point in the standard coordinate data, determine whether the coordinate point exceeds the preset normal activity area threshold; if the coordinate point exceeds the preset normal activity area threshold, mark the coordinate point as an outlier; calculate the outlier ratio based on the number of coordinate points marked as outliers; if the outlier ratio exceeds the preset outlier threshold, determine that the standard coordinate data is outlier data; filter the standard coordinate data based on the outlier detection result to obtain filtered standard coordinate data.

[0048] Specifically, after obtaining the standard coordinate data, a normal activity area threshold is preset, for example, a range of 50 meters centered on the employee workstation coordinates. Then, traverse each coordinate point in the standard coordinate data and calculate the Euclidean distance between the coordinate point and the employee workstation coordinates. If the distance exceeds 50 meters, the coordinate point is marked as an outlier. After traversing all coordinate points, the number of outliers is counted, and the proportion of outliers to the total number of coordinate points under the coordinate cluster is calculated. If the proportion of outliers exceeds the preset abnormal threshold, such as 20%, it is determined that the batch of standard coordinate data is abnormal as a whole and needs to be filtered. During filtering, the system will use the median filtering algorithm to calculate the median coordinates of all coordinate points, and then use the median coordinates as the center. The coordinate points within the normal activity area threshold range are retained, and the coordinate points outside the range are removed. Finally, the filtered standard coordinate data is obtained for subsequent standard working coordinate range generation and other processing.

[0049] Step S102: reasonably divide the standard working coordinate range of each work group according to the job attributes of the employees and the layout of the workshop. Use geometric figures such as rectangles and polygons to describe the range, and try to minimize the overlapping area between the working ranges of adjacent work groups to reduce the complexity of subsequent judgment.

[0050] Obtain the employee's job attribute information and the layout information of the workshop, and determine the standard work coordinate range of each employee based on this information. For each employee's standard work coordinate range, use geometric figures such as rectangles and polygons to describe it in combination with the above information and standard coordinate data to obtain a geometric model of the work range of the work group. Calculate the overlapping area between the geometric models of the work ranges of adjacent work groups. If the overlapping area exceeds the preset threshold, it is determined that there is an excessively large overlapping area. Adjust the geometric model of the work range of the work group through an optimization algorithm to reduce the overlapping area between the work ranges of adjacent employees, and obtain an optimized geometric model of the work range of the work group. Subsequently, based on the optimized geometric model of the work range of the work group, re-determine the standard work coordinate range of each employee and store it in the database.

[0051] Specifically, the employee's job attribute information is obtained from the human resources management system, including the department to which the employee belongs, job responsibilities, work tasks, etc., and the layout information of the workshop is extracted from the workshop layout diagram, including equipment location, material storage area, personnel channel, etc. According to the employee's job attributes and workshop layout information, such as different departments or responsibilities or tasks, different K values ​​of clustering algorithms are determined, and the initial clustering center point of the clustering algorithm is set according to the location of the equipment or material storage area or personnel channel. The K-means clustering algorithm is used to divide the coordinates of the employees into different work group coordinates. In each work group, the coordinate data of the corresponding employees are integrated into the standard work coordinates of the work group through the above clustering method, and each work group corresponds to a standard work coordinate range. For the standard work coordinate range of each work group, according to the standard work coordinates of the work group, the minimum enclosing rectangle algorithm is used to generate a rectangular geometric model, and the Graham scanning algorithm is used to generate a polygonal geometric model. The overlapping area between the geometric models of adjacent work groups is calculated. If the overlapping area exceeds 30% of the total area, it is judged that there is an excessively large overlapping area. The simulated annealing algorithm is used to optimize the geometric model of the work group. Based on the minimization of the standard working coordinate range of the work group, the standard working coordinates are set within the standard working coordinate range as a constraint. The Graham scanning algorithm is used to optimize the setting of the polygonal geometric model. By setting the initial temperature to 100, the end temperature to 0.1, and the cooling coefficient to 0.95, it is iterated 100 times, and the optimized geometric model of the work group is obtained as the working area of ​​the work group, so that the overlapping area between adjacent work groups is reduced to less than 30%. At the same time, the working area of ​​the work group is divided, and the overlapping position is regarded as the overlapping area, and the non-overlapping position is regarded as an independent area or a standard working area.

[0052] Step S103: for overlapping work groups, further subdivide the overlapping areas. Through cluster analysis of employee historical location coordinates, the activity patterns and residence time distribution characteristics of each employee in the overlapping area are obtained, and the overlapping areas are divided accordingly to meet the actual work trajectory of each employee as much as possible.

[0053] Based on the employee standard coordinate data and timestamp collected in step S101, the residence time distribution of each employee in the overlapping area is calculated to obtain the employee's activity pattern and frequency characteristics in the area; by analyzing the activity pattern and residence time characteristics, the similarities and differences of the activities of the same employee in the overlapping area are judged, and the basis for dividing the overlapping area is determined; the Graham scanning algorithm is used to subdivide the overlapping area, and the areas with similar activity patterns are used as a preliminary area division plan for the same employee; according to the actual work trajectory of the employee, the preliminary division plan is optimized and adjusted to obtain an optimized area division result, the optimized area division result is in line with the employee's work route, and the cross-overlap of the employee's work scope is reduced; the optimized area division result is visualized, and the division plan is assigned to the corresponding employee to guide the employee to work in the divided work area and improve work efficiency.

[0054] Specifically, based on the main activity range and activity center of each employee in the overlapping area, according to the clustering results, the distribution of the residence time of each employee at different locations in the overlapping area is counted. According to the statistical results in different time periods (such as different times from appearing to leaving the overlapping area), the activity patterns (distribution of residence time, including location and residence time) and frequency characteristics of employees in different time periods are obtained. By calculating the cosine similarity of employee activity patterns in different time periods and the KL divergence of residence time distribution, the similarity and difference of employee activities in the overlapping area are judged. The similarity is calculated based on two adjacent time distributions. The final similarity is the mean calculation result of each similarity, and the final KL divergence is the mean calculation result of each KL divergence. The areas with similarity greater than 8 and KL divergence less than 5 are regarded as mergeable areas as the locations where employees often move in the overlapping area, and the remaining areas need to be segmented. The Graham scanning algorithm is used to subdivide the overlapping area and segment different activity areas to obtain a preliminary area division plan. According to the actual work trajectory of employees, the simulated annealing algorithm is used to optimize the division scheme, that is, the result of the Graham scanning algorithm for the merged area. The initial temperature is set to 100, the termination temperature is set to 1, the cooling coefficient is set to 95, and it is iterated 500 times to obtain the regional division results (optimized mergeable areas) that meet the work routes of employees and the independent areas of the work area of ​​the work group as the work areas of employees. The regional division results are pushed to relevant employees in real time to guide them to work in the divided work areas and improve work efficiency and collaboration capabilities.

[0055] Step S104, when determining the area where the employee is located, first check whether the coordinate point falls within the standard working range of the employee (the working area of ​​the work group). If so, directly determine that the employee is in his or her working area; if not, further determine whether the coordinate point is in the overlapping area.

[0056] According to the pre-established standard work area data for each employee, the real-time location coordinates of the current employee are obtained; it is determined whether the location coordinates fall within the standard work area of ​​the employee. If so, it is determined that the employee is located in his standard work area; if not, the boundary data of the standard work area of ​​each employee is obtained to determine whether there is overlap between the areas; for the overlapping areas, it is determined whether the location coordinates of the employee fall within the corresponding sub-area within the overlapping area. If so, it is determined that the employee's belonging area is within the overlapping area; if the employee's belonging area still cannot be determined, the employee's historical location data is clustered and analyzed through a clustering algorithm to obtain the employee's habitual activity area; the employee's current location is matched with the habitual activity area, and if the matching degree exceeds the preset threshold, the area is determined to be the employee's current location area.

[0057] Specifically, obtain the boundary coordinate data of the employee's standard work area, and use the RayCasting algorithm to determine whether the employee's real-time location coordinates are within the standard work area. If they are within the area, mark the area as the employee's current location. If not within the standard work area, subdivide the sub-areas in the overlapping area. Determine again whether the employee's location falls within the sub-area. If so, it is considered to still belong to the employee's belonging area. If it still cannot be determined, retrieve the employee's location data for the past week, use the DBSCAN clustering algorithm for clustering, set the radius parameter eps to 5 meters, and the minimum number of samples to 50 to obtain the employee's main activity area. Calculate the Euclidean distance between the current position and the center of gravity of each activity area. If the minimum distance is less than 10 meters, the activity area is determined as the employee's current area. Finally, store the judgment results to facilitate subsequent optimization of employee behavior trajectory analysis and regional division.

[0058] Step S105: If the employee's position coordinates exceed all preset working ranges, they are marked as abnormal. The system automatically generates warning information and notifies relevant responsible personnel to verify the situation in time. At the same time, the trajectory playback and behavior analysis program are started to determine whether there are obvious abnormalities in the employee's actual work trajectory.

[0059] The real-time location coordinates of the employee are obtained and compared with the preset working range to determine whether the employee's location exceeds the preset range. If the employee's location coordinates exceed the preset working range, the employee is marked as abnormal and an early warning message is generated. According to the early warning information, the relevant responsible personnel are determined and the early warning information is pushed to the relevant responsible personnel to notify them to verify the employee's abnormal situation in time. At the same time, the employee trajectory playback program is started to obtain the historical location data of the employee within a certain time range and visualize the trajectory.

[0060] Specifically, in the real-time monitoring system, the real-time location coordinates of employees are first obtained through GPS positioning technology, and compared with the preset work area boundary data to determine whether any employee's location exceeds the preset range. For example, if the employee's coordinates exceed the boundary coordinates of the work area, the system automatically marks the employee as abnormal and generates an early warning message. This early warning message includes the employee ID, abnormal time and location coordinates. The system then pushes the early warning message to relevant responsible personnel, such as security administrators, through the internal message system so that they can verify the situation in time. At the same time, the system starts the employee trajectory playback program, extracts the employee's location data in the past 24 hours, and uses map visualization technology to display its movement trajectory. Through cluster analysis, such as the K-means algorithm, the system analyzes the employee's historical trajectory data to identify whether there are frequent deviations from the regular path. Then, the density-based DBSCAN anomaly detection algorithm is used to analyze the employee's behavior pattern and identify abnormal patterns in behavior, such as staying in a non-working area for more than the set threshold time of 30 minutes.

[0061] Step S106: Based on the above determination results, a regional determination report of the employee is generated, including basic information of the employee, the determined region, determination confidence, abnormal status, etc. The report data is automatically aggregated into the employee management database for subsequent attendance, performance evaluation, and other applications. At the same time, the regional determination results are fed back to the employee in real time to facilitate self-management and behavior optimization.

[0062] According to the employee ID, the basic information of the employee is obtained from the employee information database, such as name, work number, department, etc. The regional determination result data is integrated with the basic information of the employee to generate the regional determination report of the employee. The report content includes the basic information of the employee, the region where the determination is located, the determination confidence, abnormal status, etc. The generated employee regional determination report data is automatically summarized into the employee management database and stored in association with other data of the employee, such as attendance records, performance evaluation, etc., to facilitate subsequent data application. The data in the employee management database is analyzed and mined, and the employee behavior patterns are discovered through machine learning algorithms such as decision trees and cluster analysis to provide a basis for performance evaluation. The regional determination results are fed back to the corresponding employees in real time, and employees are informed of their regional determination status in a timely manner through mobile push notifications or emails. Employees conduct self-management and behavior optimization based on the feedback of the regional determination results. If abnormal status is found, they can be adjusted in time to improve work efficiency, optimize attendance performance, and accumulate favorable data for performance evaluation.

[0063] Specifically, the employee information database is queried through the employee ID to obtain basic information such as the employee's name, work number and department, and this information is integrated with the regional determination results to form a detailed regional determination report. The report will include the employee's basic information, the determined area, the confidence level (for example, 95%) and the abnormal status. This report is then automatically uploaded to the employee management database and stored in association with the employee's attendance records, performance evaluation and other data. In order to further analyze the employee's behavior pattern, the system uses a decision tree algorithm to analyze the employee's historical location data and work performance data to identify whether there are regular behavior patterns or abnormal patterns. For example, if an employee often appears in the work area during non-working hours, the system may mark it as a behavior that needs attention. At the same time, under the guidance of the above-mentioned work scope, relevant personnel can be prevented from entering the dangerous area of ​​​​the power equipment outside the work scope as much as possible. Finally, the regional determination results are pushed to the employee's mobile terminal in real time through the internal network. Employees can understand their own regional determination by viewing the push notification and adjust their behavior accordingly to optimize their work performance and attendance records.

[0064] 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 locating personnel in a power enterprise plant, characterized in that: include: Obtaining employee location coordinate data and preprocessing the employee location coordinate data; Integrate the preprocessed coordinate data to obtain the work group coordinate data, and obtain the work group working area according to the work group coordinate data; Obtain and obtain the residence time distribution according to the timestamp of the personnel's location coordinate data, and divide and optimize the overlapping areas of the work group's work area according to the residence time distribution to obtain the mergeable areas; The real-time location of the employee is obtained, and the real-time location is judged according to the independent area and the mergeable area of ​​the work group's work area to obtain the real-time location judgment result to realize the judgment result of whether the person is in the work area.

2. The method according to claim 1, characterized in that: The process of preprocessing employee location coordinate data includes: The employee location coordinate data is converted to a standard coordinate system, filtered and clustered. The filtering includes: performing threshold detection on the unified coordinate data through an outlier detection method, and removing the coordinate data that does not meet the threshold; the clustering analysis includes: clustering the filtered coordinate data through a DBSCAN density clustering algorithm to obtain the coordinate data of each employee corresponding to the coordinate cluster, and performing threshold detection on the coordinate data in the coordinate cluster again to obtain the preprocessed coordinate data.

3. The method according to claim 1, characterized in that The process of obtaining the work area of ​​the working group includes: The employee's job attribute information and the layout information of the workshop are obtained, and the preprocessed coordinate data are clustered into different workgroup coordinate data according to the employee's job attribute information and the layout information of the workshop by using the K-means clustering algorithm, and the workgroup coordinate data are processed by the Graham scanning algorithm to obtain a polygonal geometric model, and the polygonal geometric model is optimized according to the optimization algorithm, and the minimization of the polygonal geometric model is taken as the goal, and the overlapping area of ​​the polygonal geometric model is taken as less than or equal to a threshold as the stopping condition, so as to obtain different workgroup working areas, wherein each workgroup working area includes an overlapping area and an independent area.

4. The method according to claim 1, characterized in that The process of obtaining the residence time distribution includes: Based on the preprocessed coordinate data and corresponding timestamps of the overlapping area, the coordinate data and the corresponding timestamps are calculated to count the residence time distribution of different locations.

5. The method according to claim 1, characterized in that The process of dividing the overlapping areas of the work areas of the work groups includes: The similarity of the residence time distribution in different time periods and the KL divergence of the residence time distribution are calculated. According to the similarity calculation results and the KL divergence, the overlapping areas are segmented, and the segmentation results are optimized through the optimization algorithm to obtain the mergeable areas.

6. The method according to claim 1, characterized in that The process of determining the real-time location includes: The real-time location is judged according to the independent area. When the real-time location is in the independent area, the person is within the working range. Otherwise, the real-time location is judged according to the mergeable area. When the real-time location is in the mergeable area, the person is within the working range. Otherwise, the historical location data corresponding to the person is obtained, and cluster analysis is performed on the historical location data to obtain the regional center of gravity position. A threshold judgment is performed based on the regional center of gravity position and the real-time location. If the distance between the regional center of gravity position and the real-time location is less than the threshold, the person is within the working range. Otherwise, it is judged as not being within the working range and is an abnormal activity.

7. The method according to claim 1, characterized in that The real-time location determination result also includes: If the real-time position judgment result is out of the working range, an early warning message is generated, which includes the employee ID, abnormal time and location coordinates, and the trajectory is played back.

8. The method according to claim 1, characterized in that The real-time location determination result also includes: The historical location data of employees is analyzed through a decision tree algorithm to determine whether there are abnormal patterns, where the abnormal patterns include situations where the time spent in the work area during non-working hours is greater than a threshold.

Images