Power grid operation safety patrol risk prediction method integrated with artificial intelligence technology

Abstract

The invention discloses a power grid operation safety patrol risk prediction method fused with an artificial intelligence technology, and relates to the technical field of power grid operation safety patrol, and the method specifically comprises the following steps: carrying out the operation logic analysis of all recognized operation tasks which are executed in parallel, screening out all operation tasks with operation conflicts, and carrying out the operation logic analysis; an operation mutual interference task cluster is formed; parallel job interference feature information in the operation mutual interference task cluster is obtained and analyzed, and the task interference degree between job tasks in the operation mutual interference task cluster is evaluated; and according to an evaluation result, executing dynamic regulation and control on the job task input data set for risk prediction. According to the method, the problem of task conflict caused by operation requirements and security constraint differences of different operation tasks during parallel execution of multiple tasks in a power grid operation patrol process is solved, and the task interference degree is accurately evaluated by dynamically regulating and controlling the operation task input data set, so that the accuracy of risk prediction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation safety inspection, and in particular to a power grid operation safety inspection risk prediction method integrating artificial intelligence technology. Background Art

[0002] Power grid operation safety inspections involve systematically inspecting and recording equipment operating status, environmental safety factors, and potential fault hazards during power system operation and maintenance, particularly at key nodes such as high-voltage transmission lines, substation equipment, and distribution facilities, using manual or intelligent equipment (such as inspection robots and drones). The goal is to identify and eliminate safety hazards, ensure the safety of operators, and ensure the stable operation of the power grid system. However, traditional inspection methods suffer from information silos, delayed response times, and a reliance on empirical experience in risk identification, making it difficult to effectively predict future risks. To address this, a risk prediction method for power grid operation safety inspections that integrates artificial intelligence (AI) has emerged. Based on multi-source data (such as historical inspection records, environmental parameters, and image and video information), it utilizes AI algorithms such as machine learning and deep learning to build a risk identification and prediction model. This model enables real-time analysis of potential safety risks, trend analysis, and proactive early warning. This helps operators take preventative measures before risks occur, effectively improving inspection efficiency and reducing risks to personnel and equipment. By deeply embedding artificial intelligence technology into power grid operation scenarios, this method not only enhances the intelligence and automation capabilities of security management and control, but also lays a technical foundation for building a more forward-looking and responsive power grid security assurance system.

[0003] The existing power grid operation safety inspection risk prediction technology that integrates artificial intelligence technology mainly conducts forward-looking predictions on potential risks in the power grid operation process by building a data-driven intelligent analysis model. Its risk prediction process usually includes the following key links: First, in the data collection link, the system acquires multimodal data such as images, videos, environmental parameters (such as temperature and humidity, wind speed, electromagnetic intensity) and operator behavior trajectories of the operation site by deploying various sensors, drones, inspection robots and video surveillance; secondly, in the data preprocessing and fusion link, the system cleans, denoises, and extracts features of the collected data, and integrates structured and unstructured information to form a complete operation safety portrait; then, in the risk modeling stage, the system uses Artificial intelligence algorithms, such as convolutional neural networks (CNNs) for image recognition and long short-term memory networks (LSTMs) for processing time series data, or integrated learning models for comprehensive assessment, learn potential risk characteristic patterns from historical accident samples. Then, in the risk prediction phase, the model will analyze the deviations and changing trends between inspection data and historical data in real time, predicting high-risk events such as equipment failures, operational violations, and environmental changes. Finally, through the risk grading and early warning module, the prediction results are converted into specific risk level assessments, and combined with the geographic information system (GIS) or dispatch platform to generate visual early warning prompts, which are fed back to the operation dispatcher or on-site operator in real time, thus achieving closed-loop management of early perception, proactive intervention, and intelligent decision-making. Overall, this technology, with artificial intelligence as its core, runs through the entire process from data perception to intelligent decision-making, significantly improving the scientific and intelligent level of power grid operation safety management.

[0004] The existing technology has the following deficiencies:

[0005] During power grid inspections, when multiple tasks are executed in parallel, such as when both a clearance operation and equipment maintenance are scheduled for the same line segment, conflicting instructions can easily occur due to the different operational requirements and safety constraints of each task. For example, maintenance operations typically require a power outage, while clearance operations rely on line power to ensure the operation of the equipment. In this case, indiscriminately inputting the data of both tasks as a single batch into an AI risk prediction model will result in the model being unable to identify the task exclusion relationship between them, confusing the task context logic, and causing low-risk tasks that should be safe to be misclassified as high-risk, or high-risk tasks to remain unidentified. Existing AI-integrated power grid inspection risk prediction technologies cannot dynamically adjust the input dataset for risk prediction based on the degree of task interference when multiple tasks conflict. This results in the model being unable to decouple operational logic and identify conflicts under parallel conditions, potentially leading to biased prediction results, incorrect scheduling strategies, conflicting work plans, and even personal and equipment safety incidents.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0007] The purpose of the present invention is to provide a power grid operation safety inspection risk prediction method integrating artificial intelligence technology to solve the problems in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for predicting the risk of power grid operation safety inspections by integrating artificial intelligence technology, specifically comprising the following steps:

[0009] During the power grid inspection process, the received power grid inspection and dispatch data is analyzed to identify all the operations currently being performed in parallel within the same power equipment area;

[0010] Performing operational logic analysis on all identified parallel tasks, filtering out all tasks with operational conflicts, and forming task clusters with conflicting operations.

[0011] Obtain and analyze the interference feature information of parallel tasks in the operation-interference task cluster to evaluate the degree of task interference between tasks in the operation-interference task cluster.

[0012] Based on the evaluation results, dynamic regulation is performed on the input data set of the task used for risk prediction;

[0013] Perform risk prediction based on the regulated input data, output the risk results of each operation task, and record the risk results and corresponding operation conditions to optimize the subsequent risk prediction process.

[0014] Preferably, an operation logic analysis is performed on all the identified parallel execution tasks to screen out all the operation tasks with operation conflicts and form an operation-interference task cluster, specifically:

[0015] Based on all identified parallel tasks, the operation step information, equipment target information and safety dependency rules of each task are extracted;

[0016] Analyze the execution order relationship between tasks based on operation step information, analyze the device access relationship between tasks based on device target information, and analyze the control condition relationship between tasks based on security dependency rules;

[0017] According to the analysis results, all the tasks with operational conflicts are screened out, and these tasks are merged into a task cluster with operational interference.

[0018] Preferably, obtaining and analyzing the interference feature information of parallel tasks in the operation-interference task cluster to evaluate the degree of task interference between the tasks in the operation-interference task cluster specifically includes the following steps:

[0019] Obtaining the interference feature information of parallel operations in the task clusters that interfere with each other, and performing preprocessing after obtaining;

[0020] Extracting resource interference analysis information and step interference assessment information from the pre-processed parallel job interference feature information, and analyzing them after extraction to generate a resource conflict intensity coefficient and an operation step conflict index respectively;

[0021] Based on the generated resource conflict intensity coefficient and operation step conflict index, a task interference index is generated through weighted summation;

[0022] Determine the pre-set task interference index threshold range, and compare it with the generated task interference index after determination, and evaluate the task interference degree between each operation task in the operation mutual interference task cluster based on the comparison.

[0023] Preferably, the logic for obtaining the resource conflict intensity coefficient is as follows:

[0024] The resource interference analysis information is extracted from the pre-processed parallel job interference feature information, including the proportion of the total equipment resources allocated to each task in the operation mutual interference task cluster, the number of times the equipment resources are requested, and the number of resource access conflicts with other tasks, and are marked as A and B respectively. i 、R i and C i , A i R represents the proportion of the total equipment resources allocated to the i-th task in the operation interference task cluster, i represents the number of times the i-th task in the task cluster that interferes with each other requests device resources, C i represents the number of resource access conflicts between the i-th task and other tasks in the task cluster where i = 1, 2, 3, ..., g, where g is a positive integer;

[0025] Calculate the resource conflict intensity coefficient. The specific calculation formula is as follows:

[0026]

[0027] Where RCIC is the resource conflict intensity coefficient.

[0028] Preferably, the logic for obtaining the operation step conflict index is as follows:

[0029] The step interference evaluation information is extracted from the pre-processed parallel operation interference feature information, including the number of steps executed by each task in the operation mutual interference task cluster, the number of conflicts with other tasks in the execution steps, and the proportion of equipment resources occupied when executing the steps, and are calibrated as S i 、P i and U i , S i represents the number of steps executed by the i-th task in the task cluster where operations interfere with each other, P i represents the number of conflicts between the i-th task and other tasks in the task cluster in terms of execution steps, U i represents the proportion of device resources occupied by the i-th task in the operation interference task cluster when executing the step, i = 1, 2, 3, ..., g, where g is a positive integer;

[0030] Calculate the operation step conflict index. The specific calculation formula is as follows:

[0031]

[0032] Where OSCI is the operation step conflict index.

[0033] Preferably, based on the generated resource conflict intensity coefficient RCIC and operation step conflict index OSCI, a task interference index is generated by weighted summation. The specific calculation formula is as follows:

[0034] TII=ω1*RCIC+ω2*OSCI

[0035] Where TII is the task interference index, ω1 and ω2 are the non-zero weight coefficients of the resource conflict intensity coefficient RCIC and the operation step conflict index OSCI, respectively, and ω1+ω2=1.

[0036] Preferably, a predetermined task interference index threshold interval [TII min ,TII max ], and after determination, it is compared with the generated task interference index TII, and the task interference degree between each task in the operation mutual interference task cluster is evaluated based on the comparison. The specific comparison analysis is as follows:

[0037] If TII <TII min ,The task interference degree between each operation task in the operation mutual interference task cluster is low;

[0038] If TII min ≤TII≤TII max ,The task interference degree between each operation task in the ,operation mutual interference task cluster is moderate;

[0039] If TII>TII max,The task interference degree between each operation task in the ,operation mutual interference task cluster is the severity degree.

[0040] Preferably, according to the evaluation results, dynamic control is performed on the task input data set used for risk prediction, specifically:

[0041] If the assessment result is low, the original state of the task input data set used for risk prediction is maintained without any adjustment to the data, ensuring that the data is within the pre-set normal range for risk prediction;

[0042] If the assessment result is moderate, the task input data set used for risk prediction is weighted and adjusted to increase the weight of tasks involving high conflict in the task data;

[0043] If the assessment result is severity, the input data set of the work tasks used for risk prediction is filtered and reordered, giving priority to the data of high-interference tasks.

[0044] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0045] 1. This invention effectively solves the problem of task conflicts caused by different operational requirements and safety constraints between tasks when multiple tasks are executed in parallel. Traditional methods treat multiple tasks as a unified batch and fail to identify the exclusive relationship between tasks, resulting in low-risk tasks being misclassified as high-risk, or high-risk tasks not being identified. By analyzing the degree of task interference and dynamically adjusting the task input data set under different interference levels, this solution can accurately evaluate and distinguish conflicts between tasks in a multi-task parallel environment, effectively avoiding the conflict confusion and task misjudgment in traditional methods.

[0046] 2. This invention combines key indicators such as the resource conflict intensity coefficient and the operation step conflict index to generate a task interference index through weighted summation. Task interference is then classified and processed according to set threshold intervals. This mechanism prioritizes high-interference tasks during task data processing, ensuring that these tasks receive appropriate attention in risk prediction. The strategy of dynamically regulating the input dataset of job tasks ensures the rational allocation of resources during task scheduling and ensures more accurate and reliable risk assessment results for each task.

[0047] 3. The present invention ensures that task conflicts are effectively resolved during the risk prediction process by screening, weighting, adjusting, and prioritizing the data of high-conflict tasks. In the event of severe interference, the system can reorder and screen the input data based on the degree of conflict between tasks, giving priority to evaluating high-risk tasks. This not only improves the prediction accuracy of power grid operation safety inspections, but also provides more reliable decision-making support for the stable operation of the power grid. Through these precise interference assessments and data regulation, this technical solution significantly improves the safety of power grid operations and reduces the risk of scheduling errors and potential safety accidents caused by task conflicts. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0049] Figure 1 This is a flow chart of the power grid operation safety inspection risk prediction method that integrates artificial intelligence technology in the present invention. DETAILED DESCRIPTION

[0050] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0051] The present invention provides Figure 1 The power grid operation safety inspection risk prediction method integrating artificial intelligence technology shown in the figure specifically includes the following steps:

[0052] During the power grid inspection process, the received power grid inspection and dispatch data is analyzed to identify all the operations currently being performed in parallel within the same power equipment area;

[0053] The parallel relationship analysis of power grid operation tasks can be achieved by constructing a job scheduling data parsing algorithm. The algorithm takes the structured job information received from the scheduling platform as input, including fields such as job number, scheduling time, job type, target equipment number, power grid topology number and geographic coordinates, and maps this information to a unified time and space index system. The time overlap relationship between job tasks is identified by setting a time window sliding matching rule, and a spatial matching algorithm based on equipment attribution is used to determine whether the tasks act on the same power equipment area. Tasks with time intersection and consistent spatial attribution will be identified as parallel job tasks and further grouped into a candidate task set for subsequent analysis and processing.

[0054] When identifying all tasks currently being executed in parallel within the same power equipment area, the planned start and end times of each task are first standardized to generate a unified time interval index. The device numbers and topological area identifiers of the devices they act on are then extracted to establish a device-attribution mapping relationship for the tasks. Time interval comparisons are then used to determine the temporal intersection between tasks, and their spatial relevance is determined through device mapping relationships. When two or more tasks overlap in time and act on the same power equipment or its adjacent branches, they are considered to be in a parallel execution relationship. Ultimately, all tasks that meet the conditions are merged into the same set of parallel tasks.

[0055] Identifying parallel tasks is crucial for pre-modeling the logical relationships within the task data before risk prediction. The temporal and spatial overlap of multiple tasks often indicates potential interference between their operations. Indiscriminately incorporating these tasks as a single input batch into an AI risk prediction model makes it difficult to accurately identify the dependencies and mutually exclusive logic between tasks, leading to biased risk assessments. Proactively identifying parallel tasks during data preprocessing facilitates subsequent classification analysis and interference assessment of potentially conflicting tasks, thereby improving the reliability and accuracy of overall risk predictions.

[0056] Performing operational logic analysis on all identified parallel tasks, filtering out all tasks with operational conflicts, and forming task clusters with conflicting operations.

[0057] In this embodiment, an operation logic analysis is performed on all identified parallel execution tasks to filter out all the operation tasks with operation conflicts and form an operation-interference task cluster, specifically:

[0058] Based on all identified parallel tasks, the operation step information, equipment target information and safety dependency rules of each task are extracted;

[0059] By constructing a structured parsing algorithm for job task information, the extraction of operation step information, equipment target information, and safety dependency rules can be achieved. First, within the received dispatch platform task data, field mapping rules are used to identify metadata entries for each job task, including the job number, job type, dispatch instructions, job instruction list, power grid equipment number, and job safety control requirements. Subsequently, a rule engine performs semantic parsing and sequence analysis on the job instruction list, extracting the job execution sequence, key operation nodes, and dependency chains to form operation step information. The equipment number is matched with the power grid equipment inventory database to determine the physical equipment and topological location corresponding to the job target, which is then extracted as equipment target information. Finally, based on the job scheduling requirements and the control strategy requirements included in the safety technical specification documents, the constraint logic for the operation conditions is extracted, such as whether power outages are required, whether cordon deployment is required, and whether intervals between operations are required. These constraints are then uniformly modeled as safety dependency rules. The entire process is driven at the software level by configurable field binding and parsing rules, completing the extraction from raw task data into structured logical information for subsequent operation logic analysis.

[0060] Analyze the execution order relationship between tasks based on operation step information, analyze the device access relationship between tasks based on device target information, and analyze the control condition relationship between tasks based on security dependency rules;

[0061] By constructing a task logic relationship analysis algorithm, we can analyze the execution order, device access, and control condition relationships between tasks. First, based on the operation step information, a directed graph structure is constructed for each task's instruction sequence, extracting the temporal logic and execution dependency chains of key operation nodes. Subsequently, the node temporal logic between different tasks is cross-compared to identify task pairs with pre-dependencies, overlapping operations, or instruction conflicts, thereby determining their execution order. Second, based on the device target information, the target device number associated with each task is mapped to the power grid device topology to determine whether the tasks target the same device, the same electrical circuit, or adjacent nodes. If there is resource overlap, a device access conflict is flagged. Finally, based on the security dependency rules, the operation constraints of each task are modeled as a set of Boolean logic expressions. Through logical solution and rule matching, we analyze whether there are mutually exclusive or interfering control conditions between different tasks, such as if one task requires power outage while another requires continuous power supply. The entire analysis process involves task comparison, graph analysis, and rule calculation on an intermediate data structure, automatically generating an operation logic relationship matrix between tasks, which serves as the basis for determining operation conflicts.

[0062] According to the analysis results, all the tasks with operational conflicts are screened out, and these tasks are merged into a task cluster with operational interference.

[0063] Screening for conflicting tasks and merging them into clusters of interfering tasks can be achieved by constructing a task clustering algorithm based on a logical conflict graph. This algorithm first reads the execution order, device access, and control condition relationships between tasks analyzed in the previous step, and uses this to construct an operational logic conflict matrix between tasks, where each set of conflict relationships is represented as an edge connection between task nodes. By establishing this conflict graph structure, the system runs an aggregate traversal algorithm on the graph, identifying connections for all tasks with direct or indirect conflict paths. It then automatically merges closely interacting tasks into a preliminary set of tasks based on task relevance.

[0064] When specifically screening for conflicting operational tasks, the system determines the conflict paths for each pair of tasks based on the edge weights and conflict types between nodes in the conflict graph. When any type of valid conflict relationship exists between tasks (such as inconsistent instruction sequence, overlapping equipment resources, and conflicting control conditions), they are marked as conflicting task pairs. Subsequently, by performing a connectivity analysis in the graph, all tasks with conflicting paths are grouped into the same group, forming non-overlapping task subgraphs. Each subgraph is a cluster of mutually interfering operational tasks. This process is completed by traversing all task nodes in the graph and recording their conflicting connection relationships, ensuring that the tasks within each cluster have verifiable logical conflict links, thereby achieving structured aggregation of task conflicts.

[0065] By grouping conflicting tasks into the same mutually interfering task cluster, not only can we prevent conflict information from being misinterpreted as independent task logic in subsequent processing, but we can also provide downstream risk prediction models with clear task context boundaries, which is beneficial for the model to identify local conflict patterns and high-risk combinations. Especially in work-intensive areas or task batch execution, ignoring operational interference between tasks may lead to prediction errors or scheduling conflicts. Therefore, task screening and merging based on analysis results is a key step to improve the accuracy of artificial intelligence risk prediction and realize pre-regulatory management. It can significantly enhance the controllability of the system and the security assurance capability of task scheduling.

[0066] Obtain and analyze the interference feature information of parallel tasks in the operation-interference task cluster to evaluate the degree of task interference between tasks in the operation-interference task cluster.

[0067] In this embodiment, the parallel task interference feature information in the operation mutual interference task cluster is obtained and analyzed to evaluate the task interference degree between the various tasks in the operation mutual interference task cluster. Specifically, the following steps are included:

[0068] Obtaining the interference feature information of parallel operations in the task clusters that interfere with each other, and performing preprocessing after obtaining;

[0069] Obtaining the interference characteristic information of parallel operations in the task clusters of mutually interfering operations can be achieved by accessing the task data recorded in the power grid operation dispatching platform. First, the system needs to extract the original data related to the operation tasks from the dispatching platform, including information such as the resource request frequency of the operation tasks, resource allocation status, the type and number of task execution steps, and the number of resource access conflicts. This data is usually stored in the database in the form of structured data. The software obtains this data by calling the database interface or API and converts it into a processable format. In addition, the system can also supplement the real-time data collected by sensors and monitoring equipment with additional information related to the equipment operating status and environmental conditions, further enriching the interference characteristic information of parallel operations.

[0070] The purpose of preprocessing is to ensure that the obtained job interference characteristic information can provide high-quality data support in the subsequent analysis and modeling process. The preprocessing steps include data cleaning, denoising, standardization and other operations. First, invalid or erroneous data, such as missing values, outliers, or data that does not conform to the format, need to be removed. Secondly, standardization is performed on the values ​​of different units or ranges that may exist between different tasks to ensure that all types of data have the same dimension and comparability. Then, possible noise is smoothed, for example, by using a filtering algorithm to remove extremely fluctuating instantaneous data to ensure data stability. Finally, the system classifies and integrates the cleaned and standardized data according to task category and resource type to facilitate subsequent analysis. The preprocessed data can more accurately reflect the interference characteristics between various job tasks and provide reliable input for generating interference assessment coefficients.

[0071] Extracting resource interference analysis information and step interference assessment information from the pre-processed parallel job interference feature information, and analyzing them after extraction to generate a resource conflict intensity coefficient and an operation step conflict index respectively;

[0072] Extracting resource interference analysis information and step interference assessment information can be achieved through data analysis and feature extraction algorithms. First, within the preprocessed interference feature information, the system identifies and distinguishes between resource-related features (such as resource request frequency, resource allocation, and number of conflicts) and operational step-related features (such as the number of operational steps and the frequency of conflicts between steps). Extracting resource interference analysis information primarily relies on statistically analyzing and comparing the resources occupied by tasks, including calculating the resource utilization ratio of each task and its impact on resource access by other tasks. Step interference assessment information focuses on analyzing cross-cutting conflicts between task execution steps, particularly when multiple tasks are executed in parallel, to determine whether there are step conflicts or operational dependencies. The system classifies and clusters standardized data, using statistical analysis or correlation analysis to extract these key features and construct two types of interference analysis information. This process can be automated through machine learning models or data mining techniques to extract and categorize this information, ensuring efficient and accurate extraction.

[0073] Based on the generated resource conflict intensity coefficient and operation step conflict index, a task interference index is generated through weighted summation;

[0074] Determine the pre-set task interference index threshold range, and compare it with the generated task interference index after determination, and evaluate the task interference degree between each operation task in the operation mutual interference task cluster based on the comparison.

[0075] Determining pre-set task interference index thresholds can be achieved through data analysis and statistical methods based on historical operation data. First, the system collects and analyzes past grid operation data, particularly interference indices between different tasks (e.g., resource conflicts and operational step conflicts between tasks). These indices are then statistically analyzed, such as by calculating their mean, standard deviation, maximum, and minimum values. Based on these statistical data, the system sets interference index thresholds. Typically, these thresholds vary depending on the task type and operating environment. The system can adjust thresholds based on actual needs and experience, such as low interference (slight interference), medium interference (moderate interference), and high interference (high interference). Furthermore, the system can use machine learning algorithms to automatically optimize thresholds based on patterns in historical data, further improving the accuracy of interference index assessment. This process can be achieved through methods such as data mining, cluster analysis, or regression analysis to ensure that the threshold settings match the interference conditions encountered during actual operations.

[0076] In this embodiment, the logic for obtaining the resource conflict intensity coefficient is as follows:

[0077] The resource interference analysis information is extracted from the pre-processed parallel job interference feature information, including the proportion of the total equipment resources allocated to each task in the operation mutual interference task cluster, the number of times the equipment resources are requested, and the number of resource access conflicts with other tasks, and are marked as A and B respectively. i 、R i and C i , A i R represents the proportion of the total equipment resources allocated to the i-th task in the operation interference task cluster, i represents the number of times the i-th task in the task cluster that interferes with each other requests device resources, C i represents the number of resource access conflicts between the i-th task and other tasks in the task cluster where i = 1, 2, 3, ..., g, where g is a positive integer;

[0078] The proportion of the total equipment resources allocated to each task in the operation interference task cluster, the number of times the equipment resources are requested, and the number of resource access conflicts with other tasks can be obtained through the resource management module, operation log and monitoring system data interface in the power grid operation dispatching system. First, the proportion of the total equipment resources A i This information can be obtained by accessing the device allocation data in the resource management module. The system records the type and quantity of device resources allocated to each job task. This data is typically stored as a key-value pair of device ID and task ID. After obtaining the resource allocation information for a task, the system can calculate the proportion of resources occupied by the task, typically as the ratio of the task's resource usage to the total amount of resources allocated to that task.

[0079] Secondly, the number of times the device resources are requested R i This information can be obtained by accessing the task scheduling records in the scheduling system. Every time a job requests access to a device resource, the system records the operation and accumulates the number of requests. The software can extract each job's resource request record from the job schedule or operation log and calculate the frequency of device resource requests by counting the number of times the task requests access to the device resource.

[0080] Finally, the number of resource access conflicts with other tasks C iThis needs to be obtained by analyzing concurrent access to resources. Whenever multiple job tasks are executed in parallel and attempt to access the same device resources, the system will record a conflict event. By analyzing the device resource access log, the software can identify overlapping accesses to the same device resource between different tasks and calculate the number of resource access conflicts between each job task and other tasks. Generally, conflicts are determined based on the situation where multiple job tasks request the same device resources at the same time or in overlapping periods. The software can match and count the number of conflicts through scheduling records and device usage logs. These data are collected and processed in real time by the software system and stored in the relevant database to ensure that the resource usage, requests and conflicts of each task can be accurately tracked and calculated.

[0081] Calculate the resource conflict intensity coefficient. The specific calculation formula is as follows:

[0082]

[0083] Where RCIC is the resource conflict intensity coefficient.

[0084] The formula used to calculate the resource conflict intensity coefficient RCIC combines resource occupancy, request frequency, and conflict times to reasonably quantify the degree of resource interference between tasks. First, A in the formula i It represents the proportion of the total equipment resources allocated to each job task, which reflects the intensity of the task's occupation of equipment resources. i Multiplying, the resource request frequency of the task is also taken into account, which shows that the task with more resource requests will have a greater impact on the demand for device resources and potential conflicts. Secondly, C i Indicates the number of resource access conflicts between a task and other tasks, using the sigmoid function The number of conflicts is smoothed. Through the smoothing effect of an exponential function, when the number of conflicts is low, the interference effect between tasks is small, while when the number of conflicts is high, the impact increases rapidly. This nonlinear processing method ensures that the weight of high-conflict tasks in the overall interference coefficient is significantly increased. Ultimately, the entire formula combines the resource usage, request frequency, and conflict level of the tasks through weighted summation to derive a comprehensive resource conflict intensity coefficient, which effectively reflects the intensity of interference between tasks at the resource level.

[0085] The size of the resource conflict intensity coefficient RCIC is directly related to the degree of task interference between each job task in the task cluster that evaluates the mutual interference of operations, and the two are positively correlated. Specifically, when the RCIC value is large, it means that the job task occupies a high amount of resources and requests resources more frequently. At the same time, resource access conflicts between tasks are also more frequent, indicating that the degree of interference between tasks is high. In this case, the interference between job tasks will be more serious, which may lead to conflicts in task scheduling or excessive occupation of equipment resources, thereby affecting the efficiency and safety of power grid operations. On the contrary, when the RCIC value is small, it means that the task has a low demand for resources, the frequency of resource requests is low, and there are fewer conflicts. Therefore, the interference between tasks is small, and the coordination and stability of power grid operations are better. Through the evaluation of RCIC, the degree of resource interference between job tasks can be quantified, helping the system determine whether scheduling adjustments or optimization of the execution order of job tasks are needed to avoid potential conflicts and safety hazards.

[0086] In this embodiment, the logic for obtaining the operation step conflict index is as follows:

[0087] The step interference evaluation information is extracted from the pre-processed parallel operation interference feature information, including the number of steps executed by each task in the operation mutual interference task cluster, the number of conflicts with other tasks in the execution steps, and the proportion of equipment resources occupied when executing the steps, and are calibrated as S i 、P i and U i , S i represents the number of steps executed by the i-th task in the task cluster where operations interfere with each other, P i represents the number of conflicts between the i-th task and other tasks in the task cluster in terms of execution steps, U i represents the proportion of device resources occupied by the i-th task in the operation interference task cluster when executing the step, i = 1, 2, 3, ..., g, where g is a positive integer;

[0088] The number of steps executed by each task in the operation interference task cluster, the number of conflicts with other tasks in the execution steps, and the proportion of equipment resources occupied when executing the steps can be obtained through the data interface of the integrated job scheduling system, task execution monitoring system and equipment resource management module. First, the number of steps (S i ) can be obtained by extracting the operational steps of each task from the task execution log. Each task typically executes multiple steps according to a predetermined process. The system can analyze the task execution log records to count the total number of steps for each task. This data is generally stored in a structured form in the task management database and can be accessed by the software through calling an interface.

[0089] Secondly, the number of step conflicts (Pi ) can be obtained by monitoring the resource usage of multiple tasks in real time during concurrent execution. Conflicts arise when two or more tasks access the same device resources while executing a step. By comparing task execution time windows with resource usage, the system can determine the number of times each task conflicts with other tasks during the execution of the step. When conflicts occur, the system records the conflict event and compiles statistics for subsequent analysis.

[0090] Finally, the step resource occupancy ratio (U i ) refers to the proportion of device resources occupied by each job task during the execution of the steps. The actual utilization of device resources during task execution can be obtained through the resource management module. The system tracks the resource types and utilization amounts used by each task in each step, and then calculates the proportion of device resources occupied by each task during the execution of the step. This data is calculated by real-time monitoring of device resource usage and combining it with the task allocation information in the job scheduling system to obtain the resource utilization ratio of each task step. In this way, the system can obtain this data accurately and in real time, providing effective support for subsequent interference assessment.

[0091] Calculate the operation step conflict index. The specific calculation formula is as follows:

[0092]

[0093] Where OSCI is the operation step conflict index.

[0094] The calculation formula of the Operation Step Conflict Index (OSCI) combines the number of steps in the task, the frequency of step conflicts, and the resource occupancy ratio to reasonably evaluate the interference intensity of the steps during the task execution. Ui Use exponential operation to process the step resource usage ratio U i This is because in the task execution step, the increase in resource occupancy ratio will have a nonlinear amplification effect on the impact of other tasks. That is, when a task occupies more resources in the execution step, its interference with other tasks will increase sharply. This sharp increase can be better reflected by modeling it with an exponential function. Then, P i It represents the number of conflicts between tasks and other tasks in the execution process. It directly reflects the competition and conflict between tasks in the execution process. The higher the number of conflicts, the greater the interference intensity. Therefore, when multiplied by other factors, it can increase the impact of conflicts between tasks. Finally, ln(1+S i ) Set the number of steps S iIntroducing a logarithmic calculation effectively smooths the impact of the number of steps, preventing excessive amplification of the interference index when the task has too many steps. The logarithmic function ensures that the interference intensity decreases gradually with the number of steps, ensuring a more accurate reflection of the interference effect of the steps. By combining these factors, the final Operational Step Conflict Index (OSCI) comprehensively assesses the interference intensity between individual tasks during task execution and accurately reflects the complex interference relationships between tasks through a nonlinear mathematical model.

[0095] The size of the operation step conflict index OSCI is positively correlated with the degree of task interference between each task in the task cluster. Specifically, when the OSCI value is high, it means that the task occupies a larger proportion of resources during the execution of the step (U i The value is high), and the frequency of step conflicts between tasks (P i ) is high, or the number of operation steps of the task (S i ) indicates a high degree of mutual interference between tasks during step execution, potentially leading to resource contention, operational delays, and other issues, thereby increasing the level of interference within the entire job cluster. Conversely, a low OSCI value indicates that the task execution steps consume less resources, have fewer conflicts, and have fewer steps, indicating low levels of interference between tasks and better coordination between the tasks. Therefore, the Operation Step Conflict Index can effectively reflect the intensity of interference between tasks at the step execution level, providing an important reference for task scheduling and resource allocation, helping to reduce unnecessary conflicts and optimize the execution order of tasks.

[0096] In this embodiment, based on the generated resource conflict intensity coefficient RCIC and the operation step conflict index OSCI, a task interference index is generated by weighted summation. The specific calculation formula is as follows:

[0097] TII=ω1*RCIC+ω2*OSCI

[0098] Where TII is the task interference index, ω1 and ω2 are the non-zero weight coefficients of the resource conflict intensity coefficient RCIC and the operation step conflict index OSCI, respectively, and ω1+ω2=1.

[0099] To calculate the task interference index (TII), a weighted summation of the generated resource conflict intensity coefficient (RCIC) and the operation step conflict index (OSCI) is first performed. After extracting and calculating these two conflict indices from the task data, the system assigns appropriate non-zero weights (ω1 and ω2) to each index based on the specific requirements of the task. These weights reflect the relative importance of resource conflict and operation step conflict in task interference assessment. Typically, these weights are determined through historical data or experimental analysis and adjusted based on the nature of the task and the actual working environment. To ensure the rationality of these two weights, ω1 + ω2 = 1, meaning their sum is 1, ensuring a balanced overall interference assessment. By adjusting the weights, the system can flexibly and precisely adjust the impact of resource conflict and operation step conflict on the task interference index based on the characteristics and working conditions of different tasks, ultimately outputting an accurate task interference index (TII). These calculations can be automated through software, ensuring dynamic adaptability and efficiency across different tasks and environments.

[0100] In this embodiment, the predetermined task interference index threshold interval [TII min ,TII max ], and after determination, it is compared with the generated task interference index TII, and the task interference degree between each task in the operation mutual interference task cluster is evaluated based on the comparison. The specific comparison analysis is as follows:

[0101] If TII <TII min ,The task interference degree between each operation task in the operation mutual interference task cluster is low;

[0102] This situation indicates that there are few resource and step conflicts between tasks in the interfering task cluster, and the task execution processes are relatively independent and well-coordinated. Low interference means that the resource usage and execution steps of tasks have little impact on each other, tasks can be carried out efficiently, and the task scheduling system does not need to make frequent adjustments. This situation generally indicates stable system operation, smooth execution of tasks according to plan, and efficient resource utilization.

[0103] If TII min ≤TII≤TII max ,The task interference degree between each operation task in the ,operation mutual interference task cluster is moderate;

[0104] This indicates a certain degree of resource conflict and step interference between tasks, but not yet a severe level of interference. This suggests that some resource contention or step conflict may occur during task execution, requiring system intervention and optimization, such as adjusting the task execution order and reallocating resources, to reduce interference. While system operation may be impacted to some extent, it generally maintains basic efficiency and stability.

[0105] If TII>TII max ,The task interference degree between each operation task in the ,operation mutual interference task cluster is the severity degree.

[0106] This situation indicates significant resource and process conflicts between tasks, potentially impacting task execution. Severe interference indicates poor coordination between tasks and severe resource contention, potentially leading to task delays, equipment overloads, or operational failures. Immediate system intervention is required, potentially including rescheduling tasks, adjusting task priorities, increasing resource allocation, or suspending some tasks to prevent disruptions to the overall grid's operations and safety.

[0107] Based on the evaluation results, dynamic regulation is performed on the input data set of the task used for risk prediction;

[0108] In this embodiment, based on the evaluation results, dynamic control is performed on the task input data set used for risk prediction, specifically:

[0109] If the assessment result is low, the original state of the task input data set used for risk prediction is maintained without any adjustment to the data, ensuring that the data is within the pre-set normal range for risk prediction;

[0110] When the assessment result is low, the input dataset for the task used for risk prediction remains unchanged in its original state, allowing risk prediction analysis to be performed directly using the raw data. This process first relies on the system to automatically identify the degree of interference between tasks. If the assessment result indicates that the task interference is low, the system can determine that the raw data is sufficiently stable for subsequent risk prediction analysis. In this case, the software system automatically extracts the recorded raw task data by accessing the task information in the database. This data includes key characteristics such as the resource request frequency and execution steps of the task, and this data does not exceed the preset normal range. Due to the low interference, the raw data already provides sufficient prediction accuracy, and no additional adjustment or weighting is required. This approach avoids unnecessary calculations or processing in cases of low interference, ensuring efficient utilization of system resources, while maintaining the integrity and consistency of the task data and avoiding errors or inaccurate prediction results that may be caused by excessive intervention.

[0111] If the assessment result is moderate, the task input data set used for risk prediction is weighted and adjusted, increasing the weight of tasks involving high conflict in the task data, so that the impact of these tasks in risk prediction is more significant;

[0112] When the assessment result is moderate, weighting the task input dataset can effectively increase the influence of high-conflict tasks in risk prediction. This process first requires the system to analyze conflicts between tasks and identify those with high conflict. Typically, the system calculates the conflict level for each task based on the frequency of resource requests, overlap in step execution, and other conflict factors. When a task has a high conflict level, the system automatically assigns it a higher weight. This weighting process amplifies relevant data about conflicting tasks, such as the number of resource requests and conflict frequency, to increase the influence of high-conflict tasks. To implement this process, the software can set dynamic weights based on task conflict. The weights can be adjusted based on historical data or real-time analysis results, ensuring that tasks with high conflict receive more attention in subsequent risk predictions. This approach helps ensure that the prediction model can identify and prioritize high-risk tasks even when task interference is moderate, preventing them from potentially impacting operational safety. This approach improves prediction accuracy and reliability while also making task scheduling more scientific and rational.

[0113] If the assessment result is severe, the input data set of the work tasks used for risk prediction will be screened and reordered, giving priority to the data of high-interference tasks to ensure that key tasks are given priority in risk prediction.

[0114] When the assessment result is severity, the process of filtering and reordering the task input dataset is designed to ensure that high-interference tasks are prioritized in risk prediction. The software analyzes conflicts and interferences between tasks to identify those with high interference risk. These tasks may typically consume significant equipment resources, frequently request resources, or frequently conflict with other steps. First, the system analyzes the data of all tasks and assigns each task an interference weight based on its interference level, with higher weights indicating greater interference. The software then filters tasks based on these interference weights, prioritizing data for tasks with higher interference levels and removing or delaying data for tasks with lower interference levels. The software then reorders the task data based on their interference weights, ensuring that high-interference tasks are prioritized in risk prediction. This prioritizes their data during risk prediction, preventing potential threats to grid operations from being overlooked. This optimizes task scheduling and resource allocation, mitigating potential risks associated with high-interference tasks and ensuring the safety of tasks and the stability of system operations.

[0115] Perform risk prediction based on the regulated input data, output the risk results of each operation task, and record the risk results and corresponding operation conditions to optimize the subsequent risk prediction process.

[0116] The process of performing risk prediction based on regulated input data first relies on the regulation and optimization of task data. When the regulated input data set is used for risk prediction through an algorithmic model (such as a machine learning or deep learning model), the system analyzes the combined impact of various factors such as conflicts between tasks, resource usage, and operational steps. The software system accesses the regulated data and inputs it into the risk assessment model. The model will evaluate each task individually and output the risk results for each task, usually including a risk level or specific risk score. By predicting the risk of tasks, the system can promptly identify high-risk tasks and provide data support for scheduling and resource allocation.

[0117] Furthermore, the system records the risk results of each task and the corresponding operational status, generating a detailed risk assessment report. This record includes information such as task execution time, resource allocation, interference conditions, and predicted risk level. The purpose of this record-keeping process is to accumulate historical data and optimize subsequent risk prediction processes. The software analyzes these historical risk assessment results, identifying patterns and regularities within them, and adjusts and optimizes the parameters and algorithms of the risk prediction model to ensure more accurate risk predictions for the next time. This historical data-based optimization process improves the model's accuracy and reliability, ensuring the safety of power grid operations.

[0118] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0119] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0120] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0121] Those skilled in the art will appreciate that the units and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel 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.

[0122] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0123] Units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0124] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0125] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A power grid operation safety inspection risk prediction method integrating artificial intelligence technology is characterized by: The specific steps include: During the power grid inspection process, the received power grid inspection and dispatch data is analyzed to identify all the operations currently being performed in parallel within the same power equipment area; Performing operational logic analysis on all identified parallel tasks, filtering out all tasks with operational conflicts, and forming task clusters with conflicting operations. Obtain and analyze the interference feature information of parallel tasks in the operation-interference task cluster to evaluate the degree of task interference between tasks in the operation-interference task cluster. Based on the evaluation results, dynamic regulation is performed on the input data set of the task used for risk prediction; Perform risk prediction based on the regulated input data, output the risk results of each operation task, and record the risk results and corresponding operation conditions to optimize the subsequent risk prediction process.

2. The power grid operation safety inspection risk prediction method integrating artificial intelligence technology according to claim 1 is characterized in that: Perform an operational logic analysis on all identified parallel tasks, filter out all tasks with operational conflicts, and form task clusters with conflicting operations. Specifically: Based on all identified parallel tasks, the operation step information, equipment target information and safety dependency rules of each task are extracted; Analyze the execution order relationship between tasks based on operation step information, analyze the device access relationship between tasks based on device target information, and analyze the control condition relationship between tasks based on security dependency rules; According to the analysis results, all the tasks with operational conflicts are screened out, and these tasks are merged into a task cluster with operational interference.

3. The power grid operation safety inspection risk prediction method integrating artificial intelligence technology according to claim 2 is characterized in that: Obtain and analyze the interference feature information of parallel tasks in the operation-interference task cluster to evaluate the degree of task interference between tasks in the operation-interference task cluster. The specific steps include: Obtaining the interference feature information of parallel operations in the task clusters that interfere with each other, and performing preprocessing after obtaining; Extracting resource interference analysis information and step interference assessment information from the pre-processed parallel job interference feature information, and analyzing them after extraction to generate a resource conflict intensity coefficient and an operation step conflict index respectively; Based on the generated resource conflict intensity coefficient and operation step conflict index, a task interference index is generated through weighted summation; Determine the pre-set task interference index threshold range, and compare it with the generated task interference index after determination, and evaluate the task interference degree between each operation task in the operation mutual interference task cluster based on the comparison.

4. The power grid operation safety inspection risk prediction method integrating artificial intelligence technology according to claim 3 is characterized in that: The logic for obtaining the resource conflict intensity coefficient is as follows: The resource interference analysis information is extracted from the pre-processed parallel job interference feature information, including the proportion of the total equipment resources allocated to each task in the operation mutual interference task cluster, the number of times the equipment resources are requested, and the number of resource access conflicts with other tasks, and are marked as A and B respectively. i 、R i and C i , A i R represents the proportion of the total equipment resources allocated to the i-th task in the operation interference task cluster, i represents the number of times the i-th task in the task cluster that interferes with each other requests device resources, C i represents the number of resource access conflicts between the i-th task and other tasks in the task cluster where i = 1, 2, 3, ..., g, where g is a positive integer; Calculate the resource conflict intensity coefficient. The specific calculation formula is as follows: Where RCIC is the resource conflict intensity coefficient.

5. The power grid operation safety inspection risk prediction method integrating artificial intelligence technology according to claim 4 is characterized in that: The logic for obtaining the conflict index of the operation steps is as follows: The step interference evaluation information is extracted from the pre-processed parallel operation interference feature information, including the number of steps executed by each task in the operation mutual interference task cluster, the number of conflicts with other tasks in the execution steps, and the proportion of equipment resources occupied when executing the steps, and are calibrated as S i 、P i and U i , S i represents the number of steps executed by the i-th task in the task cluster where operations interfere with each other, P i represents the number of conflicts between the i-th task and other tasks in the task cluster in terms of execution steps, U i represents the proportion of device resources occupied by the i-th task in the operation interference task cluster when executing the step, i = 1, 2, 3, ..., g, where g is a positive integer; Calculate the operation step conflict index. The specific calculation formula is as follows: Where OSCI is the operation step conflict index.

6. The power grid operation safety inspection risk prediction method integrating artificial intelligence technology according to claim 5 is characterized in that: Based on the generated resource conflict intensity coefficient RCIC and operation step conflict index OSCI, the task interference index is generated by weighted summation. The specific calculation formula is as follows: TII=ω1*RCIC+ω2*OSCI Where TII is the task interference index, ω1 and ω2 are the non-zero weight coefficients of the resource conflict intensity coefficient RCIC and the operation step conflict index OSCI, respectively, and ω1+ω2=1.

7. The method for predicting the risk of power grid operation safety inspections integrating artificial intelligence technology according to claim 6 is characterized in that: Determine the pre-set task interference index threshold range [TII min ,TII max ], and after determination, it is compared with the generated task interference index TII, and the task interference degree between each task in the operation mutual interference task cluster is evaluated based on the comparison. The specific comparison analysis is as follows: If TII <TII min ,The task interference degree between each operation task in the operation mutual interference task cluster is low; If TII min ≤TII≤TII max ,The task interference degree between each operation task in the ,operation mutual interference task cluster is moderate; If TII>TII max ,The task interference degree between each operation task in the ,operation mutual interference task cluster is the severity degree.

8. The power grid operation safety inspection risk prediction method integrating artificial intelligence technology according to claim 7 is characterized in that: Based on the evaluation results, dynamic control is performed on the input dataset of the task used for risk prediction, specifically: If the assessment result is low, the original state of the task input data set used for risk prediction is maintained without any adjustment to the data, ensuring that the data is within the pre-set normal range for risk prediction; If the assessment result is moderate, the task input data set used for risk prediction is weighted and adjusted to increase the weight of tasks involving high conflict in the task data; If the assessment result is severity, the input data set of the work tasks used for risk prediction is filtered and reordered, giving priority to the data of high-interference tasks.

Images