Electric power engineering supervision resource allocation method and system based on abnormal event analysis

Abstract

The invention relates to the field of project supervision management, and discloses an abnormal event analysis-based electric power project supervision resource allocation method and system, and the method comprises the steps: carrying out the clustering of a plurality of pieces of abnormal event information, and obtaining a plurality of abnormal event clusters; taking minimization of processing delay of all abnormal event clusters, minimization of working duration of a single supervisor and minimization of total resource consumption as objective functions, and constructing a multi-objective optimization model; initializing the plurality of abnormal event clusters to generate an initial population; determining a target function value of each supervision resource scheduling scheme in the initial population, and performing non-dominated sorting and congestion degree calculation based on the target function values; iteratively updating the plurality of updated supervision resource scheduling schemes; when a preset iteration generation number is reached, outputting a plurality of Pareto optimal solutions; and selecting an optimal supervision resource scheduling scheme from a plurality of Pareto optimal solutions according to actual requirements. According to the invention, the supervision resource scheduling effect can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering supervision management, more specifically, to an electric power engineering supervision resource allocation method and system based on abnormal event analysis. BACKGROUND

[0002] In the field of electric power engineering supervision, abnormal event management is a key link to ensure the safety, quality and progress of the project. The existing technology usually relies on manual experience for supervision resource allocation. Supervision personnel allocate supervision resources according to event reports and manually dispatch supervision resources, such as assigning patrol personnel or equipment to the abnormal occurrence point. This method is based on historical practices or simple priority rules, lacks systematic analysis of event characteristics, and leads to low efficiency of resource allocation, making it difficult to respond to sudden abnormal events.

[0003] Existing supervision resource scheduling systems often handle each abnormal event in isolation, ignoring the similarity and relevance between abnormal events, resulting in repeated resource investment or uneven distribution. At the same time, due to the diversity of abnormal event types, uncertain occurrence time, and limited supervision resources, manual scheduling methods cannot quickly respond to changes in events, easily leading to processing delays, uneven workloads and resource waste, making it difficult for existing technology to solve the problem of dynamic scheduling of abnormal events in electric power engineering supervision. For example, when multiple abnormal events occur at the same time, supervision personnel may be over-allocated or idle, lacking a scientific multi-objective optimization mechanism to balance processing time, work duration and resource consumption, thereby affecting the overall efficiency and effectiveness of engineering supervision. SUMMARY

[0004] The purpose of the present application is to provide an electric power engineering supervision resource allocation method and system based on abnormal event analysis, which solves the technical problem of poor effect of supervision resource scheduling and achieves the technical effect of improving the effect of supervision resource scheduling.

[0005] The embodiment of the application provides a power engineering supervision resource allocation method based on abnormal event analysis, which comprises the following steps: acquiring a plurality of abnormal event information in power engineering supervision data; clustering the plurality of abnormal event information to obtain a plurality of abnormal event clusters; wherein the abnormal event information comprises an abnormal event type, an abnormal occurrence time, an abnormal severity and a required processing resource; taking minimization of processing delay of all abnormal event clusters, minimization of working time length of a single supervisor and minimization of total resource consumption as an objective function, and constructing a multi-objective optimization model; initializing an initial population for the plurality of abnormal event clusters to generate a plurality of supervision resource scheduling schemes of supervision resource allocation sequence and supervision resource allocation amount for all abnormal event clusters; wherein the multi-objective optimization model comprises a supervision resource constraint condition, and the supervision resource constraint condition comprises a skill matching degree condition of a supervisor, a geographical location proximity condition and an available time window coincidence degree condition; determining the objective function value of each supervision resource scheduling scheme in the initial population, and performing non-dominated sorting and congestion calculation based on the objective function value; reserving excellent individuals in the plurality of supervision resource scheduling schemes through selection operation, obtaining a plurality of updated supervision resource scheduling schemes through cross operation and mutation operation on the plurality of supervision resource scheduling schemes, and iteratively updating the plurality of updated supervision resource scheduling schemes; when a preset iteration number is reached, outputting a plurality of Pareto optimal solutions, and each Pareto optimal solution representing an optimized supervision resource scheduling scheme; and selecting a best supervision resource scheduling scheme from the plurality of Pareto optimal solutions according to actual demand.

[0006] In a possible implementation, the best supervision resource scheduling scheme is selected from the plurality of Pareto optimal solutions according to actual demand, which comprises the following steps: determining the processing delay time, the total resource consumption and the working time length of each worker of each Pareto optimal solution, and acquiring an urgency demand, a resource constraint demand and a personnel load balancing demand in the actual demand; wherein the urgency demand represents an abnormal processing delay limit of the project, the resource constraint demand represents a tightness of the supervision resource, and the personnel load balancing demand represents a working time constraint on each supervisor; determining the Pareto optimal solution with the minimum processing delay time as the best supervision resource scheduling scheme according to the urgency demand; determining the Pareto optimal solution with the minimum total resource consumption as the best supervision resource scheduling scheme according to the resource constraint demand; and determining the Pareto optimal solution in which the working time length of each worker meets the personnel load balancing demand as the best supervision resource scheduling scheme according to the personnel load balancing demand; wherein the best supervision resource scheduling scheme comprises a plurality of supervision resource scheduling schemes of supervision resource allocation sequence and supervision resource allocation amount for all abnormal event clusters.

[0007] In another possible implementation, the multiple abnormal event information is clustered to obtain multiple abnormal event clusters, including: obtaining progress plan data and real-time progress data in the power engineering supervision data; determining an engineering geometric structure according to engineering geometric properties, spatial properties and information properties of the progress plan data and the real-time progress data, and adding progress properties in the engineering geometric structure to obtain a progress model; wherein the progress plan data includes engineering design data, engineering node data, task duration data and dependency relationship data, the real-time progress data includes work hour data and progress supervision data, and the progress model is used to represent the geometric shape of the engineering structure, the spatial correlation between engineering tasks and the progress correlation; simulating progress delay through the progress model to determine progress delay data; determining progress delay characteristics corresponding to the progress delay data through a time series analysis algorithm; wherein the progress delay simulation is performed by dividing an engineering area, converting model parameters into numerical simulation parameters, and determining a delay influence range through a delay diffusion equation; the time series analysis algorithm includes trend decomposition, cycle identification and residual analysis on the progress delay data to determine delay intensity, propagation speed and key path influence factors as the progress delay characteristics; the multiple abnormal event information is clustered to obtain multiple abnormal event clusters; wherein the abnormal event information includes abnormal event types, abnormal occurrence times, abnormal severity, required processing resources and progress delay characteristics.

[0008] In another possible implementation, the engineering area for the progress delay simulation division includes a preset engineering area; and the multiple abnormal event clusters include multiple sub-areas in the preset engineering area respectively corresponding to abnormal event clusters and multiple sub-areas across the preset engineering area respectively corresponding to abnormal event clusters.

[0009] In another possible implementation, the method further includes: obtaining progress delay characteristics corresponding to abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme, determining a mean value of the progress delay characteristics corresponding to the abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme as an abnormal event cluster delay characteristic of each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme; and determining supervision inspection paths of the multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme according to the abnormal event cluster delay characteristics of the multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme.

[0010] In another possible implementation, the supervision inspection paths of the multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme are determined according to the abnormal event cluster delay characteristics of the multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme, including: determining an optimal inspection order as the supervision inspection paths of the multiple abnormal event clusters through a shortest path algorithm based on a mean value of the delay intensity, a mean value of the propagation speed and a mean value of the key path influence factor.

[0011] In another possible implementation, the method further includes: when the project total duration is extended according to the adjustment instruction, obtaining the remaining inspection duration of each abnormal event cluster and the remaining total inspection duration; determining a ratio of the extended project total duration to the original project total duration as a supervision adjustment ratio value; extending the remaining inspection duration of each abnormal event cluster according to the supervision adjustment ratio value, and increasing the number of supervisors assigned to each abnormal event cluster in the optimal supervision resource scheduling scheme according to the increased number of supervisors.

[0012] In another possible implementation, the method further includes: when the project total duration is shortened according to the adjustment instruction, obtaining the remaining inspection duration of each abnormal event cluster and the remaining total inspection duration; determining a ratio of the shortened project total duration to the original project total duration as a supervision adjustment ratio value; reducing the remaining inspection duration of each abnormal event cluster according to the supervision adjustment ratio value, and reducing the number of supervisors assigned to each abnormal event cluster in the optimal supervision resource scheduling scheme according to the reduced number of supervisors; wherein the number of supervisors after the proportional reduction is not less than the minimum number of supervisors corresponding to the abnormal event cluster.

[0013] In another possible implementation, the method further includes: obtaining the optimal number of supervisors of the optimal supervision resource scheduling scheme corresponding to the abnormal event cluster; determining the number of abnormal events corresponding to the abnormal event cluster, and obtaining a preset personnel ratio value corresponding to the number of abnormal events; determining a product of the optimal number of supervisors and the preset personnel ratio value as the minimum number of supervisors corresponding to the abnormal event cluster.

[0014] The embodiments of the present application also provide a power engineering supervision resource allocation system based on abnormal event analysis, which includes units for implementing the power engineering supervision resource allocation method based on abnormal event analysis.

[0015] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0016] The embodiment of the application provides a power engineering supervision resource allocation method based on abnormal event analysis, which comprises the following steps: acquiring a plurality of abnormal event information in power engineering supervision data; clustering the plurality of abnormal event information to obtain a plurality of abnormal event clusters; constructing a multi-objective optimization model with the minimum processing delay of all abnormal event clusters, the minimum working time of a single supervisor and the minimum total resource consumption as the objective function; initializing the plurality of abnormal event clusters to generate an initial population, wherein the initial population comprises a plurality of supervision resource scheduling schemes of the supervision resource allocation sequence and the supervision resource allocation amount of all abnormal event clusters; determining the objective function value of each supervision resource scheduling scheme in the initial population, and performing non-dominated sorting and congestion calculation based on the objective function value; retaining excellent individuals in the plurality of supervision resource scheduling schemes through selection operation, obtaining a plurality of updated supervision resource scheduling schemes through cross operation and mutation operation on the plurality of supervision resource scheduling schemes, and iteratively updating the plurality of updated supervision resource scheduling schemes; when the preset iteration number is reached, outputting a plurality of Pareto optimal solutions, wherein each Pareto optimal solution represents an optimized supervision resource scheduling scheme; and selecting the best supervision resource scheduling scheme according to actual needs from the plurality of Pareto optimal solutions. Through the implementation mode, the plurality of abnormal event information in the power engineering supervision data is acquired, the plurality of abnormal event clusters are formed according to similar characteristics such as abnormal event types and occurrence times, the pertinence and preliminary execution efficiency of abnormal event cluster processing are significantly improved; the multi-objective optimization model is constructed in combination with the skill matching degree of supervisors, the geographical location proximity and the coincidence degree of available time windows, so that the resource allocation takes into account multiple core requirements; high-quality scheduling schemes can be efficiently screened, and the best scheme can be selected according to actual needs, thereby realizing dynamic and accurate allocation of supervision resources. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 A flowchart of the first power engineering supervision resource allocation method based on abnormal event analysis provided by the embodiment of the application is shown in the figure.

[0019] Figure 2 A work flowchart of the first power engineering supervision resource allocation method based on abnormal event analysis provided by the embodiment of the application is shown in the figure.

[0020] Figure 3A flowchart of a second power engineering supervision resource allocation method based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 2;

[0021] Figure 4 A work flowchart of the second power engineering supervision resource allocation method based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 3;

[0022] Figure 5 A flowchart of a third power engineering supervision resource allocation method based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 4;

[0023] Figure 6 A work flowchart of the third power engineering supervision resource allocation method based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 5;

[0024] Figure 7 A flowchart of a fourth power engineering supervision resource allocation method based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 6;

[0025] Figure 8 A flowchart of a fifth power engineering supervision resource allocation method based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 7;

[0026] Figure 9 A logic structure diagram of a power engineering supervision resource allocation system based on abnormal event analysis provided by an embodiment of the present application is shown in FIG. 8. DETAILED DESCRIPTION

[0027] It should be understood that the term “includes” when used in the specification and the appended claims herein, specifies the presence of stated features, integers, steps, operations, elements, and / or components but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0028] It should also be understood that the term “and / or” when used in the specification and the appended claims herein, means any one and / or all possible combinations of one or more of the associated listed items.

[0029] As used in the specification and the appended claims herein, the term “if’ can be construed to mean “when” or “once” or “in response to determining” or “in response to detecting” depending on the context. Similarly, the phrase “if it is determined” or “if [a described condition or event] is detected” can be construed to mean “once it is determined” or “in response to determining” or “once [the described condition or event] is detected” or “in response to detecting [the described condition or event]” depending on the context.

[0030] In addition, in the description of the present application and the appended claims, the terms "first", "second", "third", etc. are used merely to distinguish descriptions and cannot be understood as indicating or implying relative importance.

[0031] Reference in the specification to "one embodiment" or "some embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrase "in one embodiment" or "in some embodiments" in various places in the specification are not necessarily all referring to the same embodiment, although it can. The terms "including," "comprising," "having" and variations thereof are meant to encompass the terms "including but not limited to."

[0032] The existing supervision resource scheduling system often handles each abnormal event in isolation, ignoring the similarity and relevance between abnormal events, resulting in repeated investment or uneven distribution of resources, thereby affecting the overall efficiency and effectiveness of project supervision.

[0033] For the above reasons, the embodiment of the present application provides a power engineering supervision resource allocation method based on abnormal event analysis. The method comprises: acquiring a plurality of abnormal event information in power engineering supervision data; clustering the plurality of abnormal event information to obtain a plurality of abnormal event clusters; constructing a multi-objective optimization model with the objective function of minimizing the processing delay of all abnormal event clusters, minimizing the working time of a single supervisor and minimizing the total resource consumption; initializing an initial population for the plurality of abnormal event clusters, the initial population comprising a plurality of supervision resource scheduling schemes of supervision resource allocation sequence and supervision resource allocation amount for all abnormal event clusters; determining the objective function value of each supervision resource scheduling scheme in the initial population, and performing non-dominated sorting and congestion calculation based on the objective function value; retaining excellent individuals in the plurality of supervision resource scheduling schemes through selection operation, and obtaining a plurality of updated supervision resource scheduling schemes through cross operation and mutation operation on the plurality of supervision resource scheduling schemes to iteratively update the plurality of updated supervision resource scheduling schemes; when the preset iteration number is reached, outputting a plurality of Pareto optimal solutions, each Pareto optimal solution representing an optimized supervision resource scheduling scheme; and selecting the best supervision resource scheduling scheme according to actual needs from the plurality of Pareto optimal solutions. Through the implementation mode, a plurality of abnormal event information in power engineering supervision data is acquired, a plurality of abnormal event clusters are formed according to similar characteristics such as abnormal event type and occurrence time, and the pertinence and preliminary execution efficiency of abnormal event cluster processing are significantly improved; a multi-objective optimization model is constructed in combination with the skill matching degree of supervisors, the geographical location proximity and the coincidence degree of available time windows, so that the resource allocation takes into account multiple core requirements; high-quality scheduling schemes can be efficiently screened, and the best scheme can be selected according to actual needs, thereby realizing dynamic and accurate allocation of supervision resources.

[0034] In some scenarios, the power engineering supervision resource allocation method based on abnormal event analysis provided by the embodiment of the present application can be applied to the supervision efficiency in power engineering construction, and is particularly suitable for construction supervision of complex projects and large-scale power engineering across regions, such as power grid construction, substation construction and other power engineering supervision, thereby improving the efficiency of power engineering supervision.

[0035] The power engineering supervision resource allocation method based on abnormal event analysis provided by the embodiment of the present application will be described in detail below with reference to specific examples.

[0036] Figure 1 The flowchart of the first power engineering supervision resource allocation method based on abnormal event analysis provided by the embodiment of the present application is shown in Figure 1 The power engineering supervision resource allocation method based on abnormal event analysis provided by the embodiment of the present application comprises S110 to S130, which will be described in detail below.

[0037] S110. Obtain information on multiple abnormal events from the power engineering supervision data. Cluster the information on multiple abnormal events to obtain multiple abnormal event clusters. The abnormal event information includes the type of abnormal event, the time of occurrence of the abnormality, the severity of the abnormality, and the processing resources required.

[0038] Figure 2 A schematic diagram of the workflow of the first power engineering supervision resource allocation method based on abnormal event analysis provided in the embodiments of this application is shown below. Figure 2 As shown, this implementation can acquire information on multiple abnormal events from power engineering supervision data. This information originates from supervision records during the power engineering construction process, covering anomalies in equipment installation, civil construction, cable laying, and other stages. Subsequently, multiple abnormal event information can be clustered based on the similarity of abnormal event type, occurrence time, severity, and required processing resources. This grouping of similar abnormal events into multiple abnormal event clusters facilitates the centralized allocation of supervision resources later.

[0039] It should be noted that the abnormal event type is a classification of the nature of the abnormal event, which can be divided into equipment-related (such as transformer wiring errors), civil engineering-related (such as substandard foundation concrete strength), and cable-related (such as damaged cable insulation), etc. Different types correspond to different handling requirements; the abnormal occurrence time is the specific time when the abnormality was recorded, such as 9:00 on March 15, 2024; the abnormality severity is the assessment of the impact of the abnormality, which can be divided into three levels: minor, moderate, and severe; the required handling resources are the personnel, tools, etc. required to handle the abnormality, such as two electrical supervisors and a multimeter required to handle transformer wiring errors.

[0040] For example, in a certain power project, the "damaged cable insulation layer" discovered at 10:00 on March 15, 2024 is a cable-related anomaly with a general severity level, requiring one electrical supervisor and an insulation tester. This is a complete example of an anomaly event information.

[0041] It should be noted that the clustering process is a grouping process based on the similarity of the features of abnormal events. A common implementation method is the K-means algorithm, which uses the type of abnormal event, the time of occurrence, the severity, and the resources required as feature vectors, and calculates the similarity and grouping.

[0042] For example, a project has three abnormal events: Event 1 is "Transformer wiring error" at 9:00 (equipment category, general, requires 2 electrical supervisors), Event 2 is "Circuit breaker installation deviation" at 9:30 (equipment category, general, requires 2 electrical supervisors), and Event 3 is "Foundation concrete strength not up to standard" at 10:00 (civil engineering category, serious, requires 3 civil engineering supervisors). After clustering, Events 1 and 2 form an equipment cluster, and Event 3 forms a civil engineering cluster. Grouping similar events together facilitates resource allocation.

[0043] S120. A multi-objective optimization model is constructed with the objective functions of minimizing the processing delay of all abnormal event clusters, minimizing the working time of individual supervisors, and minimizing total resource consumption. An initial population is generated for multiple abnormal event clusters. The initial population includes multiple supervisory resource scheduling schemes for the allocation order and amount of supervisory resources for all abnormal event clusters. The multi-objective optimization model includes supervisory resource constraints, including the skill matching degree of supervisors, geographical proximity conditions, and available time window overlap conditions.

[0044] In this implementation, minimizing the processing delay of abnormal event clusters, the working time of a single supervisor, and the total resource consumption can be taken as objectives, corresponding to processing timeliness, workload balance, and resource economy, respectively, to construct a multi-objective optimization model.

[0045] In this implementation, an initial population can be generated for the abnormal event clusters. The initial population contains multiple supervisory resource scheduling schemes, and each scheme specifies the resource allocation order and allocation amount for all abnormal event clusters.

[0046] For example, the resource allocation order could be to process the equipment cluster first and then the civil engineering cluster.

[0047] For example, the allocation could be to assign two electrical supervisors to a cluster of equipment.

[0048] It should be noted that the goal of minimizing processing latency aims to shorten the total time from the start to the completion of processing all abnormal clusters. For example, if a cluster originally required 2 hours of processing time, it can be optimized to require 1.5 hours. The goal of minimizing the working time of individual supervisors aims to minimize the working time of each supervisor, thereby balancing the working time of each supervisor and avoiding uneven workload. The goal of minimizing total resource consumption aims to reduce the total resources required to process all clusters. For example, if 5 supervisors were originally required, 4 supervisors are required after optimization.

[0049] In this implementation, the goal of minimizing the working time of a single supervisor can be measured by the variance of the working time of the supervisor.

[0050] It should be noted that the initial population generation is a process of randomly generating multiple feasible scheduling schemes. Feasible schemes must meet constraints (such as not assigning civil engineering supervision to equipment clusters). For example, given three abnormal event clusters A, B, and C, the initial population may include scheme 1 (sequence A→B→C, allocation quantities 2, 3, 1), scheme 2 (sequence B→A→C, allocation quantities 3, 2, 1), and scheme 3 (sequence C→A→B, allocation quantities 2, 3, 2). These schemes form the initial population for subsequent optimization.

[0051] It should be noted that each scheduling scheme in the initial population includes an allocation order and an allocation amount. The allocation order is the order in which clusters are processed, such as processing clusters that occurred earlier and have higher severity first; the allocation amount is the number of resources allocated to each cluster, such as allocating 2 electrical supervisors and 1 multimeter to the equipment cluster.

[0052] For example, in a certain scheme, the allocation order is "equipment cluster → cable cluster → civil engineering cluster". The equipment cluster is allocated 2 electrical supervisors and multimeters, the cable cluster is allocated 1 electrical supervisor and insulation tester, and the civil engineering cluster is allocated 3 civil engineering supervisors and concrete strength tester. This is an individual in the initial population.

[0053] In this implementation, the multi-objective optimization model also includes supervision resource constraints, which include skill matching degree, geographical proximity degree, and available time window overlap degree, to ensure that the assigned supervision personnel can effectively handle anomalies.

[0054] It should be noted that skill matching degree represents the degree of matching between the supervisor's skills and the type of anomaly, geographical proximity represents the degree of proximity between the supervisor's location and the location where the anomaly occurred, and available time window overlap represents the degree of matching between the supervisor's time and processing time.

[0055] It should be noted that the skill matching condition requires that the supervisor's skills match the type of anomaly. For example, equipment clusters require electrical supervisors. The geographical proximity condition requires that the supervisor's location be close to the location of the anomaly. For example, if the anomaly is at substation A, a supervisor from a nearby substation should be assigned. The available time window overlap condition requires that the supervisor's time and processing time match. For example, if the processing time is 10:00-12:00, a supervisor who is available at that time should be assigned. For instance, if a certain equipment cluster needs to be processed at 10:00, and the assigned supervisor, Mr. Zhang, is an electrical engineer, lives near substation A, and is available from 10:00-12:00, then all constraints are met.

[0056] S130. Determine the objective function value for each supervisory resource scheduling scheme in the initial population, and perform non-dominated sorting and crowding calculation based on the objective function value. Select superior individuals from multiple supervisory resource scheduling schemes through a selection operation. Obtain multiple updated supervisory resource scheduling schemes through crossover and mutation operations, and iteratively update these schemes. When a preset number of iterations is reached, output multiple Pareto optimal solutions, each representing an optimized supervisory resource scheduling scheme. From these multiple Pareto optimal solutions, select the best supervisory resource scheduling scheme based on actual needs.

[0057] In this implementation, after calculating the objective function value of each scheduling scheme in the initial population, non-dominated sorting and crowding calculation can be performed. Subsequently, selection operations can be used to retain superior individuals, followed by crossover and mutation operations to obtain updated scheduling schemes. Iterative updates of multiple updated supervisory resource scheduling schemes are performed using a genetic algorithm. By repeating the iterations until a preset number of generations is reached, multiple Pareto optimal solutions can be output. Each solution is an optimized scheduling scheme, and none of these schemes is superior across all objectives.

[0058] It's important to note that non-dominated sorting stratifies the options. The first stratum consists of options that are superior to any other option in all objectives. The second stratum consists of options that are dominated by the first stratum but not by other options. For example, options A (processing delay of 1 hour, minimum working time of 2 hours, resource consumption of 5), B (1.2 hours, minimum working time of 1.5 hours, resource consumption of 4.5), and C (0.9 hours, minimum working time of 2.5 hours, resource consumption of 5.5). Options A and B are not dominated by each other and belong to the first stratum; option C is dominated by option A and belongs to the second stratum.

[0059] It should be noted that crowding represents the distance between schemes in the same layer. For example, if the crowding of schemes A and B is high, it indicates that the population has good diversity and can be preserved.

[0060] It should be noted that the selection operation can use the tournament selection of genetic algorithms to select k schemes from the population and keep the best one; the crossover operation can use single-point crossover, such as exchanging the allocation order of schemes A and B to generate a new scheme; the mutation operation can randomly adjust the allocation order or amount, such as changing the allocation amount of scheme A from 2 to 3.

[0061] For example, the sequence of scheme A is anomalous event cluster A → anomalous event cluster B → anomalous event cluster C, with resource allocations of 2, 3, and 1. The sequence of scheme B is B → anomalous event cluster A → anomalous event cluster C, with resource allocations of 3, 2, and 1. After crossover, a new scheme is generated with the sequence anomalous event cluster A → anomalous event cluster B → anomalous event cluster C (allocations of 3, 2, and 1). After mutation, the allocation of scheme A is changed from 2 to 3, generating a new scheme with the sequence anomalous event cluster A → anomalous event cluster B → anomalous event cluster C, with resource allocations of 3, 3, and 1.

[0062] It should be noted that the preset iteration number is a pre-set number of iterations that controls the termination of optimization. More iterations are set for problems with high complexity. For example, in a project with 10 anomalous clusters, the preset iteration number is 80 generations. The algorithm starts from generation 1, and performs selection, crossover, and mutation in each generation until the 80th generation ends, outputting a high-quality Pareto optimal solution.

[0063] In this implementation, after outputting multiple Pareto optimal solutions, each solution is an optimized scheduling scheme. There is no absolutely optimal solution, and the best scheme can be selected according to actual needs.

[0064] It should be noted that actual needs can be specific priority objectives of the project. For example, when rushing to meet deadlines, it is necessary to shorten processing delays, so the solution with the minimum processing delay should be selected; when there are few supervisors, it is necessary to balance working hours, so the solution with the minimum working hour variance should be selected; when costs are tight, it is necessary to reduce resource consumption, so the solution with the minimum resource consumption should be selected.

[0065] This implementation method acquires information on multiple abnormal events in power engineering supervision data, and clusters them into multiple abnormal event clusters based on similar characteristics such as abnormal event type and occurrence time. This allows for centralized processing of similar abnormal events, making the allocation of supervision resources more aligned with the inherent characteristics of the events, reducing ineffective allocation, and significantly improving the targeting and initial execution efficiency of abnormal event cluster processing.

[0066] This implementation aims to minimize the processing delay of abnormal event clusters, the working time of individual supervisors, and the total resource consumption. It constructs a multi-objective optimization model by combining the skill matching degree of supervisors, geographical proximity, and overlap of available time windows. This enables resource allocation to take into account multiple core needs, achieve dynamic balance among various objectives, and significantly improve the overall rationality of the allocation scheme.

[0067] This implementation uses a genetic algorithm to perform non-dominated sorting and crowding calculation on the initial supervision resource scheduling scheme. After iterative updates through selection, crossover, and mutation operations, multiple Pareto optimal solutions are output after reaching a preset number of generations. This can efficiently screen high-quality scheduling schemes and support the selection of the best scheme according to actual needs, thereby achieving dynamic and accurate allocation of supervision resources.

[0068] In some implementations, in S130 above, the best supervisory resource scheduling scheme is selected from multiple Pareto optimal solutions according to actual needs, including S131 to S132. S131 to S132 will be explained in detail below.

[0069] S131. Determine the processing delay time, total resource consumption, and working hours of each Pareto optimal solution, and obtain the urgency requirements, resource constraint requirements, and personnel load balancing requirements in the actual needs. Among them, the urgency requirements represent the delay limit for handling project anomalies, the resource constraint requirements represent the tension of supervision resources, and the personnel load balancing requirements represent the working time constraints for each exempted supervisor.

[0070] In this implementation, when selecting the best supervision resource scheduling scheme from multiple Pareto optimal solutions based on actual needs, the processing delay time, total resource consumption, and working hours of each staff member corresponding to each Pareto optimal solution can be determined first. At the same time, the urgency requirements, resource constraint requirements, and personnel load balancing requirements in the actual needs can be obtained.

[0071] In this implementation, the urgency requirement represents the limitation of the project on the delay of abnormal handling, the resource constraint requirement represents the tension of supervision resources, and the personnel load balancing requirement represents the working time constraint of each exempted supervisor.

[0072] It should be noted that the processing delay time is the time interval from the formation of an anomaly cluster to the complete processing of that cluster. The processing delay time directly reflects the speed of anomaly event processing. For example, the processing delay time of a certain Pareto optimal solution can be the longest time from the start to the end of processing among all anomaly clusters, because the longest delay will directly affect the overall project progress.

[0073] It should be noted that the processing delay time is calculated based on the start and end processing times of each abnormal event cluster. For example, the processing delay time of the "line overload" abnormal event cluster in a power project starts at 9:00 AM and ends at 10:00 AM is 1 hour. The processing delay time of the entire Pareto optimal solution is usually the maximum delay time of all clusters.

[0074] It should be noted that total resource consumption is used to characterize the total amount of supervisory resources invested in handling abnormal event clusters, including the sum of human and material resources, reflecting the overall cost of resource use. For example, total resource consumption can be the sum of the number of supervisory personnel multiplied by the working hours, plus the number of testing equipment shifts. Specifically, weighting coefficients can be set according to resource types, and different types of resources can be quantified and added together.

[0075] It should be noted that the calculation of total resource consumption can be adjusted according to the actual situation of the project. For example, in the abnormal handling of a certain substation, the total resource consumption includes the labor cost of 3 supervisors working for 2 hours, plus the equipment cost of 1 infrared thermometer used for 1 hour. After these are quantified according to preset weights, the sum is the total resource consumption of the Pareto optimal solution.

[0076] It should be noted that each staff member's working hours are the cumulative time spent by a single staff member in processing the clusters of exceptions assigned to them, reflecting the workload of an individual. For example, if a staff member is assigned to process 3 clusters of exceptions, and each cluster takes 1 hour, 1.5 hours, and 0.5 hours respectively, then the staff member's working hours are 3 hours.

[0077] It should be noted that the working hours of each staff member need to be accurately calculated based on the processing time of each abnormal event cluster they are involved in. For example, if staff member A is responsible for two abnormal event clusters, "tower tilt" and "conductor wear", and takes 2 hours to process "tower tilt" and 1.5 hours to process "conductor wear", then staff member A's working hours are 3.5 hours, which directly reflects their workload.

[0078] It should be noted that urgency requirements are delay limits set by the project based on its own time requirements for handling anomalies, and are usually related to the importance of the project and the scope of the impact of the anomaly. For example, for an anomaly involving the main power grid line, the project may set the urgency requirement as "handling delay time not exceeding 2 hours", because an anomaly in the main line will affect the power supply of a large area and must be handled quickly.

[0079] It should be noted that the specific content of emergency requirements can be adjusted according to the engineering scenario. For example, during the peak summer season for power engineering, the power load is high and abnormal line overload may cause tripping. At this time, the emergency requirements will be more stringent, such as requiring the processing delay time to not exceed 1 hour.

[0080] It should be noted that resource constraint requirements are limitations set by the project based on the current availability of supervision resources, reflecting the degree of resource scarcity. For example, when there is a shortage of supervision personnel or testing equipment is occupied, resource constraint requirements will be more stringent, requiring total resource consumption to be minimized.

[0081] It should be noted that the resource constraint requirement can be implemented by setting an upper limit on the total resource consumption. For example, if there are only 2 available supervisors and 1 testing device for the project, the resource constraint requirement may be expressed as "the total resource consumption shall not exceed 4 hours of work for 2 people plus 2 hours of work for 1 device". The corresponding Pareto optimal solution needs to satisfy this restriction.

[0082] It should be noted that the personnel load balancing requirement is a constraint set by the project on the working hours of each non-supervisory personnel. The purpose is to prevent individual personnel from working too long hours, which could lead to fatigue and affect work efficiency or quality. For example, the personnel load balancing requirement may be set as "the working hours of each non-supervisory personnel shall not exceed 4 hours", or "the difference in working hours among all non-supervisory personnel shall not exceed 1 hour".

[0083] It should be noted that the specific content of the personnel load balancing requirement can be adjusted according to the personnel management policy of the project. For example, in the daily supervision of a certain power project, in order to ensure the rest of the personnel, the personnel load balancing requirement is set as "the daily working hours of each exempted supervisor shall not exceed 8 hours". In case of emergency, it can be appropriately relaxed, but an upper limit will still be set, such as not exceeding 10 hours.

[0084] S132. Based on urgency requirements, determine the Pareto optimal solution with the minimum processing delay time as the optimal supervision resource scheduling scheme. Based on resource constraints, determine the Pareto optimal solution with the minimum total resource consumption as the optimal supervision resource scheduling scheme. Based on personnel load balancing requirements, determine the Pareto optimal solution among all Pareto optimal solutions where the working hours of each worker meet the personnel load balancing requirements as the optimal supervision resource scheduling scheme. The optimal supervision resource scheduling scheme includes multiple supervision resource scheduling schemes for the order and amount of supervision resource allocation for all abnormal event clusters.

[0085] In this implementation, the optimal supervision resource scheduling scheme can be selected based on different actual needs: if the actual need is urgent, the Pareto optimal solution with the minimum processing delay time can be selected as the optimal scheme; if the need is resource-constrained, the Pareto optimal solution with the minimum total resource consumption can be selected; if the need is for balancing personnel load, the Pareto optimal solution that meets this need can be selected. The optimal supervision resource scheduling scheme includes the order and amount of supervision resources allocated to all abnormal event clusters.

[0086] This implementation method first clarifies the processing delay time of each Pareto optimal solution. Combined with the urgency of the actual needs that characterize the delay limit of engineering anomaly handling, the Pareto optimal solution with the smallest processing delay time is selected as the best supervision resource scheduling scheme. This can accurately meet the delay limit of engineering anomaly handling, quickly respond to and handle abnormal events, and improve the timeliness of anomaly handling.

[0087] This implementation method obtains the total resource consumption of each Pareto optimal solution, associates it with resource constraints that characterize the degree of supervision resource tension, and selects the Pareto optimal solution with the minimum total resource consumption as the best supervision resource scheduling scheme. It can adapt to different levels of supervision resource tension, effectively reduce total resource consumption, and improve the efficiency of supervision resource utilization.

[0088] This implementation method determines the working hours of each worker in each Pareto optimal solution, compares them with the personnel load balancing requirements that characterize the workers' working time constraints, and selects the Pareto optimal solution that meets these requirements as the best supervision resource scheduling scheme. This ensures that the workers' working hours meet the preset constraints, achieves balanced allocation of personnel load, and improves the overall work progress efficiency.

[0089] Figure 3 A flowchart illustrating the second method for allocating power engineering supervision resources based on anomaly event analysis provided in this application is shown below. Figure 3As shown, in some implementations, in the above S110, multiple abnormal event information is clustered to obtain multiple abnormal event clusters, including S111 to S113. S111 to S113 will be explained in detail below.

[0090] S111. Obtain the schedule plan data and real-time progress data from the power engineering supervision data. Determine the engineering geometric structure based on the engineering geometric attributes, spatial attributes, and information attributes of the schedule plan data and real-time progress data, and add progress attributes to the engineering geometric structure to obtain the progress model. The schedule plan data includes engineering design data, engineering node data, task duration data, and dependency data; the real-time progress data includes work hour data and progress supervision data. The progress model is used to represent the geometric shape of the engineering structure, the spatial relationships between engineering tasks, and the schedule relationships.

[0091] Figure 4 A schematic diagram of the workflow of the second power engineering supervision resource allocation method based on abnormal event analysis provided in the embodiments of this application is shown below. Figure 4 As shown, in this implementation, the schedule data and real-time progress data in the power engineering supervision data can be obtained. The engineering geometric structure is determined based on the engineering geometric attributes, spatial attributes and information attributes of the schedule data and real-time progress data. Progress attributes are then added to the engineering geometric structure to obtain the progress model.

[0092] It should be noted that the schedule data refers to the expected data about the progress of the power project formulated during the planning stage, such as the planned start and end times of each stage of the project and the expected duration of the tasks; the real-time progress data refers to the progress data recorded during the actual implementation of the project, such as the actual start time of each task and the amount of work completed.

[0093] For example, the schedule data could be the planned data for a power transmission line project, which states that "the foundation construction is scheduled to be completed from March 1 to March 31, 2024"; the real-time progress data could be the actual recorded data for the project, which states that "the foundation construction actually started on March 5, 2024, and 60% of the work was completed as of March 20".

[0094] For example, the schedule data can also be the preset data of "the equipment installation task is scheduled to take 15 days" in the substation project; the real-time progress data can be the work hour record data of "the task was actually completed in 12 days". By comparing the two, it can be found that the progress is ahead of schedule or delayed.

[0095] It should be noted that engineering design data refers to design information about the structure and parameters of the power engineering project generated during the design phase, such as the design dimensions of the transmission line tower foundation and the equipment layout diagram of the substation; engineering node data refers to key progress milestones in the project, such as "foundation construction completed" and "tower erection completed"; task duration data refers to the expected duration of each engineering task, such as "the conductor erection task has a duration of 10 days"; and dependency data refers to the sequential relationship between tasks, such as the logical relationship that "foundation construction must be completed before tower erection can proceed".

[0096] For example, engineering design data can be the design parameter "the design value of the concrete strength of the wind turbine foundation is C30" in a wind power project; engineering node data can be the key node "the wind turbine foundation is completed" in the project; task duration data can be the expected time "the wind turbine hoisting task duration is 5 days"; and dependency data can be the sequential relationship "the wind turbine can only be hoisted after the wind turbine foundation is completed".

[0097] For example, in a photovoltaic power station project, the engineering design data can be the design requirement that "the installation tilt angle of the photovoltaic panels is 30 degrees"; the engineering node data can be the node that "the photovoltaic panel bracket is installed"; the task duration data can be "the task duration of photovoltaic panel installation is 20 days"; and the dependency data can be the dependency that "the photovoltaic panels can only be installed after the bracket is installed". These data together constitute the basis of the schedule plan.

[0098] It should be noted that work hour data is the record of the actual time that workers spend on tasks during the project. For example, a worker works 8 hours a day for 15 days in the "foundation construction" task. Progress supervision data is the record of the supervision personnel's inspection and supervision of the project progress. For example, the progress inspection record of the "pole erection" task by the supervision personnel every week, including the number of poles and towers completed and whether it meets the planned progress.

[0099] For example, the working hours data can be the total working hours record (3×8×10=240 hours) of 3 workers working 8 hours a day for 10 consecutive days in the "conductor erection" task of a power transmission line project; the progress supervision data can be the supervision record recorded by the supervisor when inspecting the task, "As of the 5th day, 30% of the conductor erection work has been completed, which is in line with the planned progress."

[0100] For example, in a substation project, the working hours data could be the working hours record of 2 technicians working for 6 days, 7 hours a day, in the "equipment commissioning" task; the progress supervision data could be the supervision data of the supervisor for the task, such as "the progress check on the 3rd day shows that the equipment commissioning is 40% complete, which is 5% behind the plan". Through these data, the actual progress of the task can be understood.

[0101] It should be noted that engineering geometric attributes are the geometric features of the engineering structure, such as the height of transmission line towers and the diameter of the foundation; spatial attributes are the spatial relationships between different parts of the project, such as the distance between towers and the spatial relationship between substations and transmission lines; information attributes are the non-geometric and non-spatial attributes of the project, such as the construction unit, the contractor, and the budget amount.

[0102] For example, the geometric attributes of a power transmission line project can be "the tower height is 30 meters and the foundation diameter is 2 meters"; the spatial attributes can be "the distance between towers is 500 meters and the line runs in an east-west direction"; and the information attributes can be "the construction unit is a power company and the construction unit is an engineering company".

[0103] For example, the geometric attributes in a wind power project can be "the length of the wind turbine blades is 50 meters"; the spatial attributes can be "the spacing between wind turbines is 300 meters, and they are distributed on the south side of the hillside"; and the information attributes can be "the project budget is 50 million yuan". These attributes describe the characteristics of the project from different dimensions.

[0104] S112. Simulate schedule delays using a schedule model to determine schedule delay data. Utilize time series analysis algorithms to identify the schedule delay characteristics corresponding to the data. This involves dividing the project area, converting model parameters into numerical simulation parameters, and determining the scope of delay impact using delay propagation equations. The time series analysis algorithm includes trend decomposition, period identification, and residual analysis of the schedule delay data to determine delay intensity, propagation speed, and critical path impact factors as schedule delay characteristics.

[0105] In this implementation, schedule delay can be simulated using a schedule model to determine schedule delay data. Schedule delay simulation can be achieved by dividing the project area, converting model parameters into numerical simulation parameters, and determining the range of delay impact through delay diffusion equations.

[0106] In this implementation, the progress delay data can be processed using time series analysis algorithms to determine the corresponding progress delay characteristics.

[0107] In this implementation, the time series analysis algorithm includes trend decomposition, period identification, and residual analysis of the schedule delay data, thereby determining the delay intensity, propagation speed, and critical path impact factor as schedule delay features. The schedule delay features are vectorized representations of the delay intensity, propagation speed, and critical path impact factor.

[0108] It should be noted that the delay diffusion equation can be an empirical equation preset with empirical values. The delay diffusion equation is a mathematical equation used to simulate the propagation range of schedule delays in engineering. Through this equation, it is possible to calculate which subsequent tasks will be affected by the schedule delay of a certain task and the extent of the impact.

[0109] For example, suppose the "foundation construction" task in a power transmission line project is delayed by 5 days. By using the delay diffusion equation, we can input parameters such as the delay time of the task, its dependency on subsequent tasks (e.g., tower erection can only be carried out after the foundation construction is completed), and the construction period of each task. We can then calculate that the "tower erection" task will be delayed by 5 days and the "conductor erection" task will be delayed by 3 days. In this way, we can determine that the scope of the delay is the task chain of "foundation construction → tower erection → conductor erection" and the delay time of each task.

[0110] For example, in a substation project, the "equipment procurement" task was delayed by 10 days. By using the delay propagation equation and combining the dependency that "equipment installation can only be carried out after equipment procurement is completed", it was calculated that the "equipment installation" task was delayed by 10 days and the "equipment commissioning" task was delayed by 8 days. Thus, it was determined that the scope of the delay impact included the three tasks of "equipment procurement", "equipment installation" and "equipment commissioning". This allows for a clear understanding of the propagation of the delay.

[0111] It should be noted that delay intensity refers to the severity of a schedule delay, that is, the magnitude or degree of increase of the schedule delay for a certain task or project. For example, if a task is delayed by 5 days, its delay intensity is 5 days; or if the delay time of a project increases by 2 days per month, its delay intensity is the rate of "increase of 2 days per month". Propagation speed refers to the speed at which a schedule delay propagates from one task to another. For example, if a delay in one task takes 3 days to propagate to the next dependent task, its propagation speed is "3 days / task". Critical path impact factor refers to the degree of impact of a schedule delay on the critical path of the project. The critical path is the longest task chain in the project and determines the total duration of the project. The larger the value of this factor, the greater the impact of the delay on the critical path. For example, if the impact of a task delay on the critical path is 0.7, it means that the delay will cause the total duration of the critical path to increase by 70%.

[0112] For example, if the critical path of a project is "foundation construction → tower erection → conductor stringing", and the "foundation construction" is delayed by 5 days, the critical path impact factor is 0.8, then the total project duration of the critical path will increase by 4 days (5 × 0.8).

[0113] For example, the critical path of a substation project is "equipment procurement → equipment installation → equipment commissioning". If "equipment procurement" is delayed by 10 days, the critical path impact factor is 0.9, and the total duration of the critical path will increase by 9 days. This can clearly show the degree of impact of the delay on the total duration.

[0114] S113. Cluster the information of multiple abnormal events to obtain multiple abnormal event clusters. Among them, the abnormal event information includes the abnormal event type, the time of occurrence of the abnormality, the severity of the abnormality, the required processing resources, and the progress delay characteristics.

[0115] In this implementation, multiple abnormal event information can be clustered to obtain multiple abnormal event clusters. The abnormal event information includes the abnormal event type, the time of occurrence of the abnormality, the severity of the abnormality, the required processing resources, and the progress delay features obtained in the previous steps. In this way, the clustering will combine the progress delay-related features, making the features of the obtained abnormal event clusters more comprehensive.

[0116] This implementation method obtains the schedule plan data and real-time progress data from the power engineering supervision data, combines the engineering geometric attributes, spatial attributes, and information attributes, and adds the progress attribute to the engineering geometric structure to obtain the progress model. The progress delay data is determined by the progress delay simulation, and then the progress delay features are extracted by the time series analysis algorithm. These features are then incorporated into the abnormal event information for clustering, making the abnormal event clusters contain more comprehensive features and the clustering results more accurate.

[0117] This implementation method converts model parameters into numerical simulation parameters by dividing the project area, uses the delay diffusion equation to determine the scope of delay impact to simulate schedule delay, and combines the simulation results to improve abnormal event information and perform clustering. This allows for an accurate grasp of the actual impact scope and propagation of schedule delay, improving the pertinence and rationality of subsequent supervision resource allocation.

[0118] This implementation method uses time series analysis algorithms to perform trend decomposition, period identification, and residual analysis on schedule delay data, extracting schedule delay features such as delay intensity, propagation speed, and critical path impact factors. These features are then combined to complete the clustering of abnormal event information, comprehensively capturing the core characteristics of schedule delays, making abnormal event clustering more scientific, and effectively improving the efficiency of supervision resource utilization.

[0119] In some implementations, the project area for simulating schedule delays includes a pre-defined project area.

[0120] In this implementation, the engineering area divided during the progress delay simulation includes a preset engineering area. This preset engineering area is a pre-defined regional unit in the power engineering. The sub-regions within the preset engineering area are further subdivisions of the preset area, used to more accurately locate the location and impact range of abnormal events.

[0121] It should be noted that the preset engineering area is an area pre-defined during the power engineering supervision process based on the engineering design scheme, geographical layout, or functional division, and has clear boundaries and functional attributes.

[0122] For example, in overhead transmission line projects, the pre-defined project area can be a section of the line within a certain administrative region, such as "the area of ​​the 3rd section of the 110kV transmission line within XX city"; in substation projects, the pre-defined project area can be a functionally defined area such as the main transformer area or the high-voltage switchgear area. Such pre-defined area division can focus on engineering activities within a specific scope, providing a clear spatial basis for subsequent progress delay simulation and abnormal event clustering.

[0123] In some implementations, multiple exception event clusters include multiple sub-regions within a preset project area corresponding to exception event clusters, and multiple sub-regions spanning the preset project area corresponding to exception event clusters.

[0124] In this implementation, when the project area includes a preset project area, the corresponding multiple exception event clusters specifically include the exception event clusters corresponding to each of the multiple sub-regions within the preset project area, as well as the exception event clusters corresponding to each of the multiple sub-regions spanning the preset project area.

[0125] It should be noted that the multiple sub-regions within the preset engineering area are further subdivisions of the preset engineering area, with each sub-region having a more specific function or a smaller geographical scope.

[0126] For example, when the preset engineering area is "XX substation main transformer area", the sub-areas can be the main transformer 1 bay area, the main transformer 2 bay area, the cooling system layout area, etc., corresponding to different equipment or functional modules within the main transformer area; within the preset engineering area of ​​a transmission line (a certain section of tower group), the sub-areas can be the construction area corresponding to each tower, such as "tower No. 101 foundation construction area" and "tower No. 102 erection construction area", etc. Such subdivision can more accurately locate the location of abnormal events, allowing abnormal event clusters to directly correspond to specific engineering parts.

[0127] This implementation method divides the engineering area into preset engineering areas during schedule delay simulation. The clustered abnormal event clusters cover the abnormal event clusters corresponding to each sub-area within the preset engineering area and the abnormal event clusters corresponding to each sub-area across the preset engineering area. This makes the abnormal event clustering more consistent with the actual regional distribution of the project and improves the accuracy of abnormal event classification. Furthermore, resource allocation can accurately match the abnormal event handling needs of different regional types, reduce resource mismatch, and improve the rationality and efficiency of supervision resource scheduling.

[0128] With this implementation, after the abnormal event cluster is divided into sub-regions within the preset engineering area and cross-region sub-regions, the scheduling scheme in the initial population is more targeted. During the iteration process, the crossover and mutation operations are continuously optimized to adapt to the processing characteristics of abnormal events in different regions, so that the optimal supervision resource scheduling scheme can better balance processing delay, working time and total resource consumption.

[0129] Figure 5 A flowchart illustrating the third method for allocating power engineering supervision resources based on abnormal event analysis provided in this application is shown below. Figure 5 As shown, in some implementations, the above method also includes S210 to S220, which will be described in detail below.

[0130] S210. Obtain the progress delay characteristics of abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme, and determine the mean of the progress delay characteristics of abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme, as the abnormal event cluster delay characteristics of each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme.

[0131] Figure 6 A schematic diagram of the workflow of the third power engineering supervision resource allocation method based on abnormal event analysis provided in the embodiments of this application is shown below. Figure 6 As shown, in this implementation, the progress delay characteristics of each abnormal event in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme can be obtained, where the progress delay characteristics include delay intensity, propagation speed and critical path impact factor.

[0132] For each cluster of anomalous events, the mean delay intensity, mean propagation speed, and mean critical path impact factor of all anomalous events within the cluster can be calculated. The combination of these three means is used as the anomalous event cluster delay characteristic, thereby quantifying the overall delay characteristics of each anomalous event cluster and providing a basis for subsequent supervision and inspection path planning.

[0133] It should be noted that the calculation process for the delay feature of an anomaly event cluster can be followed as follows: When an anomaly event cluster contains 3 anomalies, the delay intensity of the first anomaly event is 0.8, the propagation speed is 2 meters / hour, and the critical path impact factor is 0.9; the delay intensity of the second is 0.7, the propagation speed is 3 meters / hour, and the critical path impact factor is 0.85; the delay intensity of the third is 0.9, the propagation speed is 2.5 meters / hour, and the critical path impact factor is 0.95. The mean of each feature can be calculated: the mean delay intensity is (0.8+0.7+0.9) / 3=0.8, the mean propagation speed is (2+3+2.5) / 3=2.5 meters / hour, and the mean critical path impact factor is (0.9+0.85+0.95) / 3=0.9. This combination of means is the anomaly event cluster delay feature.

[0134] S220. Based on the delay characteristics of multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme, determine the supervision inspection path for multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme.

[0135] In this implementation, the inspection path can be determined based on the delay characteristics of multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme. Then, by combining the average delay intensity, average propagation speed, and average critical path influence factor of the abnormal event clusters, the shortest path algorithm can be used to plan the optimal inspection sequence, allowing supervisors to inspect each abnormal event cluster in this order, thereby improving the targeting and efficiency of the inspection.

[0136] It should be noted that the process of determining the optimal inspection order based on the mean delay intensity, mean propagation speed, and mean critical path impact factor can be determined as follows: Each cluster of anomalous events is treated as a node in the path planning. The path weight of a node is determined by the three means: the higher the mean delay intensity, the higher the weight; the faster the mean propagation speed, the higher the weight; and the higher the mean critical path impact factor, the higher the weight. Dijkstra's algorithm is used to calculate the shortest path from the supervisor's starting point to each node. Nodes with higher weights are prioritized for inclusion in the path. For example, given three clusters of anomalous events A, B, and C, where A has a mean delay intensity of 0.9, a propagation speed of 2.8 m / h, and a critical path impact factor of 0.95; B has three means of 0.7, 2.0, and 0.8; and C has 0.85, 2.5, and 0.9. Therefore, A has the highest weight, followed by C, and B has the lowest. The optimal inspection order can be determined as A→C→B, which prioritizes processing anomalous event clusters with severe delays, fast propagation, and impact on the critical path.

[0137] This implementation method, after determining the optimal supervision resource scheduling scheme, calculates the delay characteristics of abnormal event clusters and plans the supervision inspection path based on these characteristics, thereby improving the supervision workflow. It extends the allocation of power engineering supervision resources from scheme determination to inspection path planning, making the entire supervision process more systematic and further optimizing the overall effectiveness and operability of supervision work.

[0138] This implementation method obtains the progress delay characteristics of abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme, calculates the mean of these progress delay characteristics to obtain the delay characteristics of the abnormal event cluster, and determines the mean of delay intensity, propagation speed and critical path impact factor. By quantifying the overall delay characteristics of the abnormal event cluster through the mean, the delay characteristics of the abnormal event cluster are made more representative, providing an accurate and comprehensive basis for subsequent supervision and inspection path planning.

[0139] In some implementations, in S220 above, the supervision and inspection path of multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme is determined based on the delay characteristics of the multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme. This includes: determining the optimal inspection order based on the average delay intensity, average propagation speed, and average critical path influence factor through the shortest path algorithm, which serves as the supervision and inspection path for multiple abnormal event clusters.

[0140] In this implementation, when determining the supervision and inspection path based on the delay characteristics of multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme, the optimal inspection order can be determined using a shortest path algorithm based on the average delay intensity, average propagation speed, and average critical path impact factor of each abnormal event cluster. This optimal inspection order is the supervision and inspection path for multiple abnormal event clusters. The delay characteristics of an abnormal event cluster are the average of the progress delay characteristics of all abnormal events within each cluster, encompassing the average delay intensity, average propagation speed, and average critical path impact factor. These averages comprehensively reflect the overall delay characteristics of each abnormal event cluster, providing a precise basis for planning the inspection order.

[0141] It should be noted that the application logic of the shortest path algorithm is to treat each cluster of abnormal events as a node in a graph structure. The edge weight between nodes is calculated by combining the differences in the delay features of two abnormal event clusters. The smaller the feature difference, the lower the edge weight, which means that the cost of switching from one cluster to another for inspection is lower.

[0142] For example, Dijkstra's algorithm can be used to start with the first cluster of anomalous events to be checked, calculate the shortest path from that starting point to all other clusters of anomalous events, and then determine the optimal checking order that covers all nodes based on this path information; alternatively, the Floyd-Warshall algorithm can be used to calculate the shortest path between all pairs of nodes, and more comprehensively traverse possible order combinations.

[0143] It should be noted that, to better align the inspection order with actual engineering needs, different weights can be assigned to different delay features, adjusting their priority in influencing the inspection order. For example, if the current project focuses more on clusters of anomalous events with high delay intensity, a higher weight can be assigned to the mean delay intensity. In this case, the edge weight between nodes with large differences in the mean delay intensity will be higher, and the algorithm will prioritize clusters of anomalous events with high mean delay intensity at the beginning of the inspection order.

[0144] For example, there are three abnormal event clusters A, B, and C. The average latency intensity of A is 0.9, that of B is 0.7, and that of C is 0.5. By calculating the weighted shortest path algorithm, the possible inspection order is A→B→C. This can prioritize the processing of clusters with high latency intensity and improve the targeting of the inspection.

[0145] This implementation combines the average delay intensity, average propagation speed, and average critical path impact factor of each abnormal event cluster, and uses the shortest path algorithm to derive the optimal inspection sequence to form the supervision and inspection path. This ensures that the supervision and inspection sequence closely matches the delay characteristics of the abnormal event clusters, effectively reducing path redundancy and significantly improving the overall efficiency of supervision and inspection. It also makes the allocation of supervision resources in the inspection process highly compatible with the characteristics of abnormal event clusters, reducing unnecessary resource consumption and improving the rationality of supervision resource utilization.

[0146] Figure 7 A flowchart illustrating the fourth method for allocating power engineering supervision resources based on abnormal event analysis provided in this application is shown below. Figure 7 As shown, in some implementations, the above method also includes S310 to S320, which will be described in detail below.

[0147] S310. When the total project duration is extended according to the adjustment instruction, obtain the remaining inspection time and the remaining total inspection time for each abnormal event cluster. Determine the ratio of the extended total project duration to the original total project duration as the supervisor's adjustment ratio value.

[0148] In this implementation, when the total project duration is extended according to the adjustment instruction, the time required for the inspection work that has not yet been completed for each abnormal event cluster can be obtained first, which is taken as the remaining inspection time for each abnormal event cluster; at the same time, the sum of the remaining inspection times of all abnormal event clusters can be calculated to obtain the total remaining inspection time.

[0149] In this implementation, after obtaining the extended total project duration (i.e., the updated total project time corresponding to the adjustment instruction) and the original total project duration, the extended total project duration can be divided by the original total project duration, and the result is used as the supervision adjustment ratio value.

[0150] S320. Extend the remaining inspection time for each abnormal event cluster according to the supervision adjustment ratio, and allocate the number of supervision personnel corresponding to each abnormal event cluster according to the optimal supervision resource scheduling plan.

[0151] In this implementation, after obtaining the supervision adjustment ratio, the remaining inspection time for each abnormal event cluster can be extended according to this ratio. For example, the original remaining inspection time for each abnormal event cluster can be multiplied by the supervision adjustment ratio to obtain the adjusted remaining inspection time. At the same time, according to the requirements corresponding to the supervision adjustment ratio, the number of supervision personnel allocated to each abnormal event cluster in the optimal supervision resource scheduling scheme can be increased, so that the inspection work of each abnormal event cluster can be carried out more fully within the extended project period.

[0152] It should be noted that the extended total project duration is the total time required to complete the entire power project after the adjustment instruction is updated, and the remaining total inspection time is the sum of the time that the inspection work of all current abnormal event clusters has not been completed; the supervisor's adjustment ratio quantifies the adjustment range of the project duration extension on the remaining inspection work. The larger the ratio, the higher the degree of adjustment required.

[0153] For example, when extending the remaining inspection time for each cluster of abnormal events according to the supervisor's adjustment ratio, the remaining inspection time can be multiplied by the supervisor's adjustment ratio to obtain the extended remaining inspection time. Similarly, the number of supervisors assigned can be adjusted by multiplying by the supervisor's adjustment ratio.

[0154] This implementation method calculates the ratio of the extended total project duration to the original total project duration as the supervision adjustment ratio. The remaining inspection time for each cluster of abnormal events is then extended proportionally. Simultaneously, the number of supervisory personnel allocated to each cluster of abnormal events in the optimal supervision resource scheduling plan is increased. This ensures that the allocation of supervision resources adapts to the adjustment of the total project duration, enabling dynamic adjustment of resource allocation as the project duration changes. This improves the flexibility and adaptability of resource allocation, avoids insufficient inspection time or personnel due to project duration extension, ensures meticulous and orderly inspection work, and enhances the rationality and sufficiency of supervision inspection execution.

[0155] Figure 8 A flowchart illustrating the fifth method for allocating power engineering supervision resources based on abnormal event analysis provided in this application is shown below. Figure 8As shown, in some implementations, the method also includes S330 to S340, which will be explained in detail below.

[0156] S330. When the total project duration is shortened according to the adjustment instruction, obtain the remaining inspection time and the remaining total inspection time for each abnormal event cluster. Determine the ratio of the shortened total project duration to the original total project duration as the supervisor's adjustment ratio value.

[0157] In this implementation, when the total project duration is shortened according to the adjustment instruction, the remaining inspection time for each abnormal event cluster can be obtained. At the same time, the remaining inspection time of all abnormal event clusters is summed to obtain the total remaining inspection time. This allows for a clear understanding of the remaining time for the current supervision and inspection work, providing accurate data support for subsequent adjustments.

[0158] In this implementation method, the ratio of the shortened total project duration to the original total project duration can be determined as the adjustment ratio value for supervision, wherein the remaining total inspection time should be less than the shortened total project duration.

[0159] For example, if the shortened total project duration is 80 hours and the original total project duration was 100 hours, then the supervisor's adjustment ratio is 80 divided by 100, which equals 0.8. This ratio quantifies the tightness of the remaining inspection time relative to the shortened project duration, providing a clear basis for subsequent time and personnel adjustments.

[0160] S340. Reduce the remaining inspection time for each abnormal event cluster according to the adjustment ratio, and allocate the number of supervisors corresponding to each abnormal event cluster according to the optimal supervisor resource scheduling plan. The number of supervisors after the proportional reduction shall not be less than the minimum number of supervisors corresponding to the abnormal event cluster.

[0161] In this implementation, the remaining inspection time for each abnormal event cluster can be reduced according to the supervisor's adjustment ratio. For example, if the original remaining inspection time for an abnormal event cluster is 25 hours, and the supervisor's adjustment ratio is 0.8, the adjusted remaining inspection time can be 25 multiplied by 0.8, which equals 20 hours.

[0162] At the same time, the number of supervisors assigned to the abnormal event cluster should be reduced according to this ratio. For example, if 6 supervisors were originally assigned, the number may be reduced to 5. However, the number of personnel after the reduction should not be less than the minimum number of supervisors required for the abnormal event cluster. For example, if the minimum is 3, then there should be no less than 3. This can both adapt to the shortened construction period requirements and ensure that there are enough personnel to complete the inspection work.

[0163] This implementation method calculates the ratio of the remaining total inspection time to the shortened total project duration as the supervision adjustment ratio. The remaining inspection time for each cluster of abnormal events is then reduced according to this ratio. Simultaneously, the number of supervisors allocated to each cluster of abnormal events in the optimal supervision resource scheduling scheme is reduced. This ensures that resource allocation is precisely adapted to the shortened project duration, avoids resource redundancy, and improves the utilization efficiency of power engineering supervision resources. It responds to the need for shortened project duration while ensuring the basic human resources required for supervision work, avoiding supervision gaps due to excessive personnel reduction, and maintaining the quality standards of supervision work.

[0164] In some implementations, the above method also includes S350 to S360, which are described in detail below.

[0165] S350. Obtain the optimal number of supervisory personnel for the best supervisory resource scheduling plan corresponding to the abnormal event cluster. Determine the number of abnormal events corresponding to the abnormal event cluster, and obtain the preset personnel ratio value corresponding to the number of abnormal events.

[0166] In this implementation, the optimal number of supervisors for the best supervisory resource scheduling scheme corresponding to the abnormal event cluster can be obtained first. This number is obtained through the iteration of the multi-objective optimization model and is the optimal personnel configuration that adapts to the processing needs of the abnormal event cluster.

[0167] For example, for a cluster of abnormal events involving line faults, the optimal number of supervisors in the best supervisory resource scheduling scheme is 4 people. This is the optimal value determined after comprehensively considering delays, resource consumption, and personnel load.

[0168] In this implementation, the number of abnormal events corresponding to the abnormal event cluster can be determined, that is, the total number of abnormal events contained in the cluster, and then the preset personnel ratio value corresponding to this number of abnormal events can be obtained.

[0169] For example, when there are 8 equipment abnormal events in an abnormal event cluster, the preset personnel ratio value can be 0.5; when there are 12 cable abnormal events in the cluster, the preset personnel ratio value can be 0.4.

[0170] S360. Determine the product of the optimal number of supervisors and the preset personnel ratio as the minimum number of supervisors corresponding to the abnormal event cluster.

[0171] In this implementation, the product of the optimal number of supervisors and the preset personnel ratio is calculated, and this product is used as the minimum number of supervisors corresponding to the abnormal event cluster.

[0172] For example, if the optimal number of supervisors is 4 and the preset ratio is 0.5, then the minimum number of supervisors is 2; if the optimal number of supervisors is 5 and the preset ratio is 0.4, then the minimum number is 2.

[0173] It should be noted that the preset personnel ratio corresponding to the number of abnormal events can be a pre-set ratio coefficient based on the number of abnormal events in the abnormal event cluster. The setting of the preset personnel ratio corresponding to the number of abnormal events is based on historical data of power engineering supervision, reflecting the correspondence between the number of abnormal events and the minimum required personnel configuration ratio. For example, the more abnormal events there are, the lower the preset personnel ratio may be, but it will ensure that the number of personnel after the product can meet the basic processing needs.

[0174] For example, when there are 3 abnormal events, the preset personnel ratio can be 0.8, because the number of events is small but each event requires more detailed inspection; when there are 18 abnormal events, the preset personnel ratio can be 0.3, because the number of events is large but basic supervision work can be covered through reasonable division of labor.

[0175] It should be noted that the purpose of the preset personnel ratio is to link the optimal number of supervisors with the scale of abnormal events, so that the minimum number of supervisors is based on both the optimal resource allocation plan and the actual number of abnormal events.

[0176] For example, when the number of abnormal events increases, even if the optimal number of personnel increases, the reduction of the preset ratio can prevent the minimum number of personnel from growing excessively; when the number of abnormal events decreases, the increase of the preset ratio can ensure that the minimum number of personnel is sufficient to handle each event.

[0177] This implementation method obtains the optimal number of supervisory personnel for the best supervisory resource scheduling scheme corresponding to an anomaly event cluster, determines the number of anomalies in the cluster and obtains the corresponding preset personnel ratio, and calculates the product of the two as the minimum number of supervisory personnel for the cluster. This ensures that the minimum number of supervisory personnel matches the scale of the anomalies and the optimal resource allocation, improving the scientific nature and adaptability of resource adjustments. It also avoids excessive reduction of supervisory personnel due to shortened project timelines, ensuring that each anomaly event cluster has sufficient personnel to carry out supervisory work, and guaranteeing the effectiveness of anomaly handling and the quality of supervisory work.

[0178] This application also provides a power engineering supervision resource allocation system based on abnormal event analysis, including a unit for implementing the above-described power engineering supervision resource allocation method based on abnormal event analysis.

[0179] Figure 9 A schematic diagram of the logical structure of a power engineering supervision resource allocation system based on abnormal event analysis is provided for an embodiment of this application, as shown below. Figure 9As shown, the system 1 of this embodiment includes a processing unit 11, a storage unit 12, and a transceiver unit 13. The processing unit 11 is used to process data, the storage unit 12 is used to store data, and the transceiver unit 13 is used to send and receive data. The processing unit 11, the storage unit 12, and the transceiver unit 13 cooperate with each other to implement the above-described method. The beneficial effects of the embodiments of this application have been described in the above-described method and will not be repeated here.

[0180] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0181] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0182] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0183] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0184] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0185] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for allocating power engineering supervision resources based on abnormal event analysis, characterized in that, The method includes: Acquire information on multiple abnormal events from power engineering supervision data; cluster the information on multiple abnormal events to obtain multiple abnormal event clusters; among them, the abnormal event information includes the type of abnormal event, the time of occurrence of the abnormality, the severity of the abnormality, and the required processing resources; A multi-objective optimization model is constructed with the objective functions of minimizing the processing delay of all abnormal event clusters, minimizing the working time of individual supervisors, and minimizing total resource consumption. An initial population is generated for multiple abnormal event clusters. The initial population includes multiple supervision resource scheduling schemes for the order and amount of supervision resource allocation for all abnormal event clusters. The multi-objective optimization model includes supervision resource constraints, which include the skill matching degree of supervisors, geographical proximity, and overlap of available time windows. The objective function value of each supervisory resource scheduling scheme in the initial population is determined, and non-dominated sorting and crowding calculation are performed based on the objective function value. Superior individuals from multiple supervisory resource scheduling schemes are retained through selection operations. Multiple updated supervisory resource scheduling schemes are obtained through crossover and mutation operations, and these schemes are iteratively updated. When a preset number of iterations is reached, multiple Pareto optimal solutions are output, each representing an optimized supervisory resource scheduling scheme. From these multiple Pareto optimal solutions, the best supervisory resource scheduling scheme is selected based on actual needs.

2. The method according to claim 1, characterized in that, From multiple Pareto optimal solutions, the best supervision resource scheduling scheme is selected based on actual needs, including: Determine the processing delay time, total resource consumption, and working hours of each Pareto optimal solution to obtain the urgency requirements, resource constraints, and personnel load balancing requirements in the actual needs. Among them, the urgency requirements represent the delay limit for handling abnormalities in the project, the resource constraints represent the tension of supervision resources, and the personnel load balancing requirements represent the working time constraints for each exempted supervisor. Based on urgency requirements, the Pareto optimal solution with the minimum processing delay time is determined as the optimal supervision resource scheduling scheme; based on resource constraint requirements, the Pareto optimal solution with the minimum total resource consumption is determined as the optimal supervision resource scheduling scheme; based on personnel load balancing requirements, the Pareto optimal solution among all Pareto optimal solutions where the working hours of each worker meet the personnel load balancing requirements is determined as the optimal supervision resource scheduling scheme; among these, the optimal supervision resource scheduling scheme includes multiple supervision resource scheduling schemes for the order and amount of supervision resource allocation for all abnormal event clusters.

3. The method according to claim 2, characterized in that, Clustering multiple anomalous event information results in multiple anomalous event clusters, including: Acquire schedule planning data and real-time progress data from power engineering supervision data; determine the engineering geometric structure based on the engineering geometric attributes, spatial attributes, and information attributes of the schedule planning data and real-time progress data, and add progress attributes to the engineering geometric structure to obtain a progress model; wherein, the schedule planning data includes engineering design data, engineering node data, task duration data, and dependency relationship data, and the real-time progress data includes work hour data and progress supervision data, and the progress model is used to represent the geometric shape of the engineering structure, the spatial relationship between engineering tasks, and the progress relationship; Schedule delays are simulated using a schedule model to determine schedule delay data. Time series analysis algorithms are then used to determine the schedule delay characteristics corresponding to the schedule delay data. Specifically, schedule delay simulation is performed by dividing the project area, converting model parameters into numerical simulation parameters, and determining the scope of delay impact through delay propagation equations. The time series analysis algorithm includes trend decomposition, period identification, and residual analysis of the schedule delay data to determine delay intensity, propagation speed, and critical path impact factors as schedule delay characteristics. Multiple abnormal event information is clustered to obtain multiple abnormal event clusters; among them, the abnormal event information includes abnormal event type, abnormal occurrence time, abnormal severity, required processing resources, and progress delay characteristics.

4. The method according to claim 3, characterized in that, The engineering area for simulating schedule delays includes a preset engineering area; multiple abnormal event clusters include multiple sub-regions within the preset engineering area corresponding to abnormal event clusters, and multiple sub-regions across the preset engineering area corresponding to abnormal event clusters.

5. The method according to claim 4, characterized in that, The method further includes: Obtain the progress delay characteristics of abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme, and determine the mean of the progress delay characteristics of abnormal events in each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme, as the abnormal event cluster delay characteristics of each abnormal event cluster corresponding to the optimal supervision resource scheduling scheme. Based on the delay characteristics of multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme, the supervision inspection paths for multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme are determined.

6. The method according to claim 5, characterized in that, Based on the delay characteristics of multiple abnormal event clusters corresponding to the optimal supervision resource scheduling scheme, the supervision inspection paths for these clusters are determined, including: Based on the mean delay intensity, mean propagation speed, and mean critical path impact factor, the optimal inspection sequence is determined using the shortest path algorithm, serving as the supervisory inspection path for multiple clusters of abnormal events.

7. The method according to claim 6, characterized in that, The method further includes: When the total project duration is extended according to the adjustment instruction, obtain the remaining inspection time and the remaining total inspection time for each cluster of abnormal events; determine the ratio of the extended total project duration to the original total project duration as the supervision adjustment ratio value; The remaining inspection time for each abnormal event cluster is extended according to the adjustment ratio of the supervision, and the number of supervision personnel allocated to each abnormal event cluster is increased according to the optimal supervision resource scheduling plan.

8. The method according to claim 7, characterized in that, The method further includes: When the total project duration is shortened according to the adjustment instruction, obtain the remaining inspection time and the remaining total inspection time for each abnormal event cluster; determine the ratio of the shortened total project duration to the original total project duration, and use it as the supervision adjustment ratio value; The remaining inspection time for each abnormal event cluster is reduced according to the adjustment ratio of the supervision, and the number of supervision personnel corresponding to each abnormal event cluster is allocated according to the reduction of the optimal supervision resource scheduling plan; wherein the number of supervision personnel after the reduction ratio is not less than the minimum number of supervision personnel corresponding to the abnormal event cluster.

9. The method according to claim 8, characterized in that, The method further includes: Obtain the optimal number of supervisors for the best supervisory resource scheduling plan corresponding to the abnormal event cluster; determine the number of abnormal events corresponding to the abnormal event cluster, and obtain the preset personnel ratio value corresponding to the number of abnormal events; The product of the optimal number of supervisors and the preset personnel ratio is used as the minimum number of supervisors corresponding to the abnormal event cluster.

10. A power engineering supervision resource allocation system based on abnormal event analysis, characterized in that, Includes units for implementing the method of any one of claims 1 to 9.

Images