A Smart Analysis Method and System for Employee Points Evaluation

Abstract

This invention provides an intelligent analysis method and system for employee point evaluation, relating to the field of employee score management technology. The method includes the following steps: acquiring real-time change data of key power grid operating parameters; identifying non-standard operating behaviors performed by dispatchers; tracking the actual changes in key power grid operating parameters after the occurrence of non-standard operating behaviors to obtain the actual recovery path; comparing the actual recovery path with the expected recovery path and calculating the contribution score of the non-standard operating behavior; evaluating the non-standard operating behavior; and determining the operational nature of the non-standard operating behavior and calculating the corresponding dispatcher's employee points based on the contribution score and evaluation results. The method of this invention aims to solve the problem that existing technologies cannot fairly evaluate dispatchers' non-standard operating behaviors, enabling objective and accurate evaluation of dispatchers' non-standard operating behaviors, fair calculation of employee points, and thus incentivizing innovative operations and improving the safe operation of the power grid.

Description

Technical Field

[0001] This invention relates to the field of employee rating management technology, and more specifically, to an intelligent analysis method and system for employee point evaluation. Background Technology

[0002] In power system dispatching and operation management, traditional employee performance evaluation systems have significant limitations. Existing technologies mainly rely on quantitative indicators based on standard operating procedures for assessment, which makes it difficult to accurately reflect the dispatcher's true professional capabilities under complex power grid conditions. Especially when emergency faults occur in the power grid, dispatchers often need to make emergency response decisions that go beyond standard procedures based on the actual situation on site, and existing evaluation systems cannot effectively identify and quantify the actual value of such non-standard operating behaviors.

[0003] While the intelligent points-based evaluation systems commonly used by power companies demonstrate good applicability in routine operation and maintenance, they reveal significant shortcomings when applied to technology-intensive departments such as dispatch and control centers. These systems typically only recognize standardized operations that conform to preset rules, lacking an effective evaluation mechanism for innovative measures implemented by dispatchers based on professional judgment. When non-standard operations taken by dispatchers involve complex scenarios such as multi-operation coordination and physical parameter interaction, existing systems cannot accurately predict the resulting power grid state evolution path, nor can they objectively evaluate their actual contribution to system stability.

[0004] A more prominent problem is that existing evaluation systems suffer from inherent bias in judging non-standard operating behaviors. These systems often automatically categorize deviations from standard procedures as abnormal behavior, ignoring their positive effects under specific grid operating conditions. This evaluation bias makes it difficult for innovative dispatchers to receive fair evaluations, impacting employee motivation and potentially suppressing innovative operational practices that benefit grid safety. Furthermore, current technologies lack the ability to comprehensively assess the short-term disturbances and long-term benefits caused by non-standard operations, and cannot quantify the physical interaction effects between different operations, further limiting the accuracy and fairness of evaluation results.

[0005] There is currently no effective technical solution to the above problems. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent analysis method and system for employee point evaluation, which aims to solve the problem that existing technologies cannot fairly evaluate the non-standard operating behavior of dispatchers. It can objectively and accurately evaluate the non-standard operating behavior of dispatchers, fairly calculate employee points, and is conducive to incentivizing innovative operations and improving the safe operation of the power grid.

[0007] In a first aspect, the present invention provides an intelligent analysis method for employee points evaluation, comprising the following steps:

[0008] S1. Obtain real-time change data of key power grid operating parameters;

[0009] S2. Identify non-standard operating behaviors performed by the dispatcher; the non-standard operating behaviors are instructions or adjustments that do not fall within the preset standard operating procedures.

[0010] S3. For the power grid fault scenario when the non-standard operating behavior occurs, refer to the power grid recovery performance under standard operating procedures in historical fault events to generate an expected recovery path;

[0011] S4. Track the actual changes in key power grid operating parameters after the occurrence of the non-standard operating behavior to obtain the actual recovery path;

[0012] S5. Compare the actual recovery path with the expected recovery path, and when it is determined that the actual recovery path is better than the expected recovery path, calculate the contribution score of the non-standard operating behavior to the improvement of power grid stability and efficiency;

[0013] S6. Evaluate the short-term disturbances caused by the aforementioned non-standard operating behaviors and their long-term positive effects, and obtain the evaluation results;

[0014] S7. Based on the contribution score and the evaluation result, determine the operational nature of the non-standard operation behavior, and calculate the corresponding dispatcher's employee points based on the determined operational nature.

[0015] The intelligent analysis method for employee point evaluation provided by this invention has the advantages of being able to objectively and accurately evaluate the non-standard operating behavior of dispatchers, fairly calculate employee points, incentivize innovative operations, and improve the safe operation of the power grid.

[0016] Secondly, the present invention provides an intelligent analysis system for employee points evaluation, comprising:

[0017] The acquisition module is used to acquire real-time change data of key operating parameters of the power grid;

[0018] The identification module is used to identify non-standard operating behaviors performed by the dispatcher; the non-standard operating behaviors are instructions or adjustments that do not fall within the preset standard operating procedures.

[0019] The generation module is used to generate an expected recovery path for the power grid fault scenario when the non-standard operating behavior occurs, by referring to the power grid recovery performance under standard operating procedures in historical fault events.

[0020] The monitoring module is used to track the actual changes in key power grid operating parameters after the non-standard operating behavior occurs, and to obtain the actual recovery path;

[0021] The comparison module is used to compare the actual recovery path with the expected recovery path, and when it is determined that the actual recovery path is better than the expected recovery path, calculate the contribution score of the non-standard operating behavior to the improvement of power grid stability and efficiency;

[0022] The evaluation module is used to evaluate the short-term disturbances caused by the non-standard operating behavior and its long-term positive impacts, and to obtain the evaluation results.

[0023] The judgment module is used to determine the operational nature of the non-standard operation behavior based on the contribution score and the evaluation result, and to calculate the employee points of the corresponding dispatcher based on the determined operational nature.

[0024] As can be seen from the above, the intelligent analysis method for employee point evaluation provided by this invention solves the problem that existing technologies cannot fairly evaluate dispatchers' non-standard operating behaviors by acquiring key power grid operating parameter data, identifying non-standard operating behaviors, generating expected recovery paths, tracking actual recovery paths, comparing paths and calculating contribution scores, assessing short-term disturbances and long-term impacts, and judging the nature of operations to calculate employee points. It has the advantages of being able to objectively and accurately evaluate dispatchers' non-standard operating behaviors, fairly calculate employee points, incentivize innovative operations, and improve the safe operation of the power grid.

[0025] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0026] Figure 1 This is a flowchart of an intelligent analysis method for employee points evaluation provided in an embodiment of the present invention.

[0027] Figure 2 This is a schematic diagram of a structural design for an intelligent analysis system for employee points evaluation provided in an embodiment of the present invention.

[0028] Label Explanation:

[0029] 100. Acquisition Module; 200. Identification Module; 300. Generation Module; 400. Monitoring Module; 500. Comparison Module; 600. Evaluation Module; 700. Judgment Module. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0031] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0032] In the employee performance evaluation process at power dispatch control centers, existing intelligent analysis systems fail to accurately identify and reward the contributions of dispatchers' non-standard operational behaviors to grid stability and efficiency. Non-standard operational behaviors refer to instructions or adjustments made by dispatchers based on professional judgment when they deviate from preset standard operating procedures. Because the system relies on historical training data and preset rules for evaluation, when dispatchers perform non-standard operations not recorded in historical data, the system is prone to misjudging them as abnormal behavior or negative events, resulting in performance evaluations that do not accurately reflect the actual value of the operations. Consequently, the performance evaluation results deviate from the principle of fairness in human resource management, failing to effectively incentivize dispatchers to make expert-level decisions in complex and emergency situations, and consequently affecting the system's adaptability to high-tech, complex, and high-risk business environments.

[0033] For example, in an incident where a regional power grid experienced severe convective weather, causing multiple transmission lines to trip due to lightning strikes, dispatcher B, through analysis of the overall power flow, anticipated the risk of the fault spreading to their area. Without clear procedural support, they implemented preventative load shedding operations on non-critical loads. This operation successfully prevented the fault from escalating, creating a stability margin for grid recovery. However, the performance evaluation system, based on the correlation between "load shedding" and negative events in historical data, flagged this operation as abnormal behavior and assigned a lower score. The system only recognized dispatcher A's standard operating instructions, awarding them a high score, while dispatcher B's non-standard operation was not correctly evaluated. This resulted in their ranking failing to reflect their actual contribution, causing a discrepancy between the evaluation results and the expert review conclusions.

[0034] If the aforementioned problems are not addressed, the points-based evaluation system will be unable to adapt to the high-tech, complex, and high-risk operational environment of power dispatch centers. Misjudgments of non-standard operations may lead to the underestimation of valuable expert decisions, diminishing dispatchers' incentive to take innovative preventative measures in emergency situations. Furthermore, this could affect the power grid's rapid recovery capability in the event of a fault, increasing system operational risks. In the long term, the unfairness of the points mechanism will undermine the effectiveness of human resource management, hindering the maintenance of the dispatch team's professional level and emergency response capabilities, thus posing a potential threat to the safe and stable operation of the power grid.

[0035] For reference, see the appendix. Figure 1 This invention provides an intelligent analysis method for employee points evaluation, comprising the following steps:

[0036] S1. Obtain real-time change data of key power grid operating parameters to reflect the current operating status of the power grid;

[0037] S2. Identify non-standard operating behaviors performed by the dispatcher; non-standard operating behaviors are instructions or adjustments that are not within the preset standard operating procedures, and record a snapshot of the power grid status when the non-standard operating behavior occurs;

[0038] S3. For grid fault scenarios where non-standard operating behaviors occur, generate expected recovery paths by referring to the grid recovery performance of historical fault events under standard operating procedures.

[0039] S4. Track the actual changes in key power grid operating parameters after non-standard operating behaviors occur to obtain the actual recovery path;

[0040] S5. Compare the actual recovery path with the expected recovery path, and when the actual recovery path is determined to be better than the expected recovery path, calculate the contribution score of non-standard operating behavior to the improvement of grid stability and efficiency;

[0041] S6. Assess the short-term disturbances caused by non-standard operating procedures and their long-term positive impacts to obtain the assessment results;

[0042] S7. Based on the contribution score and evaluation results, determine the operational nature of non-standard operating behavior, and calculate the corresponding dispatcher's employee points based on the determined operational nature.

[0043] For ease of understanding, the following explains some key terms in this embodiment:

[0044] Key power grid operating parameters refer to the core physical quantities reflecting the power grid's operating status, including but not limited to voltage, frequency, power flow distribution, load conditions, and various alarm information. These parameters are connected to the power dispatch automation system (SCADA / EMS) and the wide-area measurement system (WAMS) via data interfaces to acquire real-time data on voltage, current, frequency, active power, reactive power, circuit breaker status, and relay protection operation information from various nodes in the power grid (such as substations and power plants). The data acquisition frequency can be set according to the importance of the parameters; for example, key parameters such as voltage and frequency can be acquired at the millisecond level, while parameters such as load and power flow can be acquired at the second level. This raw data is cleaned, denoised, and formatted by a data preprocessing unit before being stored in a high-performance time-series database, such as InfluxDB or TimescaleDB. A dedicated data analysis module continuously monitors these parameters and calculates real-time stability indices of the power grid (e.g., voltage stability index, frequency deviation index) to reflect the current operating status of the power grid.

[0045] Non-standard operating procedures refer to instructions or adjustments executed by the dispatcher that are not within the preset standard operating procedures. The system maintains a "standard operating procedure library," which contains all approved standard operating sequences, instruction templates, and parameter thresholds for different grid fault types and operating states. This library can be managed and executed through a rule engine (e.g., based on Drools or a custom rule parser). All dispatcher operating instructions (such as circuit breaker closing / opening, generator starting / stopping, load adjustment, etc.) are recorded in real time through the dispatcher's operating log system. An "operation comparison module" continuously compares the dispatcher's actual operating sequences with the contents of the "standard operating procedure library." If an operating sequence is found to be inconsistent with any standard procedure, or if a critical operation is performed without corresponding work orders or contingency plans (e.g., proactively reducing load in a non-emergency situation), the operation will be marked as a "non-standard operation."

[0046] Grid status snapshot: refers to the instantaneous recording of key operating parameters of the power grid when a non-standard operation occurs. The system immediately records the detailed information of the operation, including operator identity, operation instructions, and operation time, and extracts snapshots of key power grid parameters before and after the operation from the time-series database to form a "non-standard operation event package".

[0047] Expected Recovery Path: For grid fault scenarios involving non-standard operating procedures, this path is generated by referencing historical fault events and grid recovery performance under standard operating procedures. It outlines the time required for the grid to stabilize after a fault, and the reasonable trend of grid state changes during this period. The system maintains a "Past Event Experience Summary Library," which stores detailed records of numerous past grid fault events, including fault type, occurrence time, initial grid state, execution of standard operating procedures, and the final grid recovery trajectory (e.g., voltage, frequency, and power flow recovery curves). When a new grid fault occurs, a "Scenario Matching Module" analyzes the current fault type, severity, affected area, and initial grid operating state, searching the "Past Event Experience Summary Library" for the most similar historical events. The system then synthesizes the average recovery time, voltage / frequency recovery curves, and load recovery curves of these similar historical events under strict adherence to standard operating procedures to generate an "Expected Recovery Path" as a benchmark. This path is obtained through statistical summarization and trend prediction of the recovery performance of multiple similar events, for example, through weighted averaging or trend line fitting.

[0048] Actual recovery path: refers to the actual changes in key power grid operating parameters after a non-standard operation occurs. After a non-standard operation is identified, the system continuously retrieves real-time data of key power grid operating parameters from the time-series database and plots it as an "actual recovery path" curve.

[0049] Contribution Score: This refers to the quantitative assessment result of the improvement in grid stability and efficiency brought about by non-standard operating behaviors. When the "Path Comparison Module" finds that the "Actual Recovery Path" is significantly better than the "Expected Recovery Path" in one or more of the above indicators, the system will activate the "Contribution Quantification Module". The contribution value will be calculated using the following formula: Contribution Score = (W_Time * (Expected Recovery Time - Actual Recovery Time) / Expected Recovery Time) + (W_Stability * (Expected Stability - Actual Stability) / Expected Stability) + (W_Guarantee Additional Guarantee Coefficient). Wherein, W_Time, W_Stability, and W_Guarantee are preset weighting coefficients based on the severity of the fault and the priority of grid operation. The additional guarantee coefficient is a bonus item set based on avoiding secondary faults, reducing load shedding, etc. For example, if a potential secondary fault is avoided, the additional guarantee coefficient can be set to 1.5. This reverse derivation can accurately identify and quantify the value brought about by expert-level decisions based on experience judgment and prediction that go beyond standard operating procedures.

[0050] Assessment Results: This refers to a comprehensive evaluation of the short-term disturbances caused by non-standard operating procedures and their long-term positive impacts. A "property discrimination module" analyzes the fluctuations in grid parameters over a very short period (e.g., 1-5 minutes) after a non-standard operation, quantifying its "short-term disturbance value" (e.g., voltage drop amplitude, peak frequency deviation). Simultaneously, this module combines the "contribution score" calculated in step five to assess its "long-term positive impact."

[0051] Operational nature: refers to the category of non-standard operational behavior determined based on contribution score and evaluation results, such as "benevolent optimization", "mistake operation" or "malicious sabotage".

[0052] This application proposes an intelligent analysis method for employee point evaluation, which aims to solve the problem that existing employee point evaluation systems cannot accurately identify and reward the contributions of dispatchers' non-standard operating behaviors to the improvement of power grid stability and efficiency in power dispatch centers.

[0053] First, the method involves acquiring real-time change data of key power grid operating parameters to reflect the current operating status of the grid. The system continuously collects real-time change data of key power grid operating parameters, including but not limited to voltage, frequency, power flow distribution, load conditions, and various alarm information. This data forms the basis for assessing the "health status" of the power grid. For example, by connecting to power dispatch automation systems (SCADA / EMS) and wide-area measurement systems (WAMS), real-time data such as voltage, current, frequency, active power, reactive power, circuit breaker status, and relay protection operation information of various nodes in the power grid (such as substations and power plants) can be acquired. The data acquisition frequency can be set according to the importance of the parameters; for example, key parameters such as voltage and frequency can be acquired at the millisecond level, while parameters such as load and power flow can be acquired at the second level. This raw data is cleaned, denoised, and formatted by a data preprocessing unit before being stored in a high-performance time-series database, such as InfluxDB or TimescaleDB. A dedicated data analysis module continuously monitors these parameters and calculates real-time stability indices of the power grid (e.g., voltage stability index, frequency deviation index) to reflect the current operating status of the grid. This high-frequency, multi-dimensional data acquisition is designed to provide sufficiently detailed information for subsequent analysis when the power dispatch and control center faces sudden, complex, and high-risk power grid fault scenarios.

[0054] Secondly, the method includes identifying non-standard operating behaviors performed by the dispatcher, which are instructions or adjustments not included in the preset standard operating procedures, and recording a snapshot of the power grid state at the time the non-standard operating behavior occurs. When the dispatcher executes any instruction or adjustment not included in the preset standard operating procedures, the system immediately identifies and records these "non-standard operations." The recorded content includes the time of the operation, the specific content, and a snapshot of the power grid state at the time of the operation. For example, the system maintains a "standard operating procedure library," which contains all approved standard operating sequences, operating instruction templates, and parameter thresholds for different power grid fault types and operating states. This procedure library can be managed and executed by a rule engine (e.g., based on Drools or a custom rule parser). All operation instructions of the dispatcher (such as circuit breaker closing / opening, generator starting / stopping, load adjustment, etc.) are recorded in real time through the dispatcher operation log system. An "operation comparison module" continuously compares the dispatcher's actual operation sequence with the contents of the "standard operating procedure library." If an operation sequence is found to deviate from any standard procedure, or if a critical operation is performed without corresponding work orders or contingency plans (e.g., proactively reducing load in a non-emergency situation), the operation will be marked as a "non-standard operation." The system will immediately record detailed information about the operation, including operator identity, operation instructions, and operation time, and extract snapshots of key power grid parameters before and after the operation from the time-series database to form a "non-standard operation event package." This identification mechanism is designed to capture expert-level decision-making behaviors that go beyond standard operating procedures and are based on experience and prediction.

[0055] Furthermore, the method includes generating a projected recovery path for the grid fault scenario occurring during the non-standard operating behavior, referencing the grid recovery performance under standard operating procedures in historical fault events. For the current grid fault scenario, the system references similar past fault events and combines these events with the grid recovery performance under strict adherence to standard operating procedures to generate a "projected recovery path." This path represents the time required for the grid to recover from a fault to stability under normal operation, as well as the reasonable trend of grid state changes during that period. For example, the system maintains a "past event experience summary library," which stores detailed records of numerous past grid fault events, including fault type, occurrence time, initial grid state, execution status of standard operating procedures, and the final grid recovery trajectory (such as voltage, frequency, and power flow recovery curves). When a new grid fault occurs, a "scenario matching module" analyzes the current fault type, severity, affected area, and initial grid operating state, searching the "past event experience summary library" for several historical events most similar to the current scenario. Then, the system synthesizes data from these similar historical events, including the average recovery time, voltage / frequency recovery curve, and load recovery curve of the power grid under strict adherence to standard operating procedures, to generate an "expected recovery path" as a benchmark. This path does not simply replicate a single historical event, but rather statistically summarizes and predicts trends from the recovery performance of multiple similar events, for example, through weighted averaging or trend line fitting. This approach aims to provide an objective reference point for measuring the actual effectiveness of non-standard operations in the face of sudden, complex, and high-risk power grid failure scenarios encountered by power dispatch and control centers.

[0056] Furthermore, the method includes tracking the actual changes in key power grid operating parameters after the non-standard operation occurs to obtain the actual recovery path. The system closely tracks the actual changes in key power grid operating parameters after the non-standard operation occurs, thereby depicting an "actual recovery path". For example, after a non-standard operation is identified, the system continuously obtains real-time data of key power grid operating parameters from the time-series database and plots it as an "actual recovery path" curve.

[0057] Furthermore, the method includes comparing the actual recovery path with the expected recovery path, and calculating the contribution score of the non-standard operation to the improvement of grid stability and efficiency when the actual recovery path is determined to be superior to the expected recovery path. Subsequently, the system performs a detailed comparison between this "actual recovery path" and the previously set "expected recovery path." For example, a "path comparison module" compares this "actual recovery path" with the "expected recovery path" generated in the third step in real time. The comparison indicators include: a. Recovery time: the difference between the actual time required to reach a stable state and the expected time. b. Stability level: the comparison between the fluctuation amplitude of parameters such as voltage and frequency during the recovery process and the expected fluctuation amplitude. For example, calculating the area difference between the actual curve and the expected curve, or the root mean square error (RMSE). c. Additional guarantees for grid stability: for example, whether secondary faults were avoided during the recovery process, or whether the load reduction was lower than expected while achieving the same recovery effect. Through the quantitative comparison of these indicators, the system can objectively assess the specific impact of non-standard operations on the grid recovery process. If the "actual recovery path" significantly outperforms the "expected recovery path" in terms of grid stabilization speed, recovery efficiency, or additional guarantees of grid stability (e.g., avoidance of secondary faults, load shedding less than expected but with significant effect), the system will determine that this non-standard operation has made a positive contribution to the overall stability and efficiency of the grid. The contribution value will be quantified based on the degree to which the actual path surpasses the expected path. When the "path comparison module" finds that the "actual recovery path" is significantly better than the "expected recovery path" in any one or more of the above indicators, the system will activate the "contribution quantification module." The criterion for "significantly better" can be set as follows: the recovery time is shortened by more than a certain percentage (e.g., 5%), or the improvement in additional guarantees of grid stability (such as voltage deviation integral, frequency deviation integral) exceeds a certain threshold.

[0058] Furthermore, the method includes assessing the short-term disturbances caused by the non-standard operation and their long-term positive impacts to obtain an assessment result. The system simultaneously assesses the short-term disturbances caused by the non-standard operation to the grid state and their long-term positive impacts. For example, a "property discrimination module" analyzes the fluctuations of grid parameters within a very short time (e.g., 1-5 minutes) after the non-standard operation occurs, quantifying its "short-term disturbance value" (e.g., voltage drop amplitude, frequency deviation peak). Simultaneously, this module combines the "contribution score" calculated in step five to assess its "long-term positive impact." The discrimination logic is as follows: a. If the "contribution score" is positive and reaches a preset threshold, and the "short-term disturbance value" is lower than the preset threshold, it is judged as "benevolent optimization." b. If the "contribution score" is negative, or the "contribution score" is positive but does not reach the threshold, and the "short-term disturbance value" is higher than the preset threshold, it is judged as "erroneous operation." c. If the operation causes a significant deterioration in the grid state and has no positive contribution, it is judged as "malicious sabotage." For example, a brief load reduction (moderate short-term disturbance) that quickly leads to optimized power flow distribution in the grid and avoids line overload tripping (significant long-term positive impact) is considered a "benevolent optimization." This judgment mechanism is designed to effectively distinguish between malicious violations and unconventional operations aimed at solving problems in the context of sudden, complex, and high-risk grid fault scenarios faced by power dispatch and control centers.

[0059] Finally, the method includes determining the operational nature of the non-standard operation based on the contribution score and the evaluation results, and calculating the corresponding dispatcher's employee points based on the determined operational nature. Based on the assessed contribution value and operational nature, the system will add the corresponding points to the dispatcher's total score, ensuring that employees who make decisions that go beyond procedures but are beneficial to the overall power grid at critical moments receive fair recognition and incentives. For example, a "points summary module" will sum the "contribution score" (if it is "benevolent optimization") or "negative score" (if it is "misoperation" or "malicious sabotage") determined in step six with the dispatcher's regular points in other areas (such as attendance, standard task completion, training participation, etc.). The final points will be updated in real time to the employee's personal performance file and displayed through the company's internal performance management system. The system will also provide a detailed points report explaining how the contribution value of the non-standard operation was calculated, including its specific impact on the speed and degree of power grid stability, to enhance the transparency and persuasiveness of the evaluation results.

[0060] The core technical concept of this solution lies in its shift from merely focusing on whether dispatchers strictly adhere to pre-set procedures to evaluating the actual impact of these operations on the overall operation of the power grid. It establishes a power grid recovery benchmark based on past experience and then correlates non-standard operations performed by dispatchers in emergency situations with the degree of improvement in the actual power grid recovery path. If the actual recovery effect is significantly better than the benchmark, the system will work backward to deduce the positive contribution of the non-standard operation to improving power grid stability and efficiency, and quantify its value accordingly. Simultaneously, it meticulously analyzes the short-term impact and long-term benefits of operations, thereby accurately distinguishing between innovative operations performed to solve problems and genuine errors or violations.

[0061] The following example will provide a more detailed explanation of the above technical solution:

[0062] Suppose a rare severe convective weather event hits a regional power grid, causing multiple transmission lines to trip due to lightning strikes. Dispatcher A, on duty, strictly followed the accident handling procedures, meticulously isolating the faults, determining their nature, attempting line reclosing, and deploying repair teams according to the pre-planned steps. Each of his actions was clear and standardized, with complete records kept in the dispatch log, and the entire process proceeded smoothly. Meanwhile, Dispatcher B, responsible for a neighboring area, although less affected, anticipated the risk of the fault spreading to his area and potentially triggering a chain reaction through analysis of the overall network power flow. Without explicit procedural support, Dispatcher B decisively implemented a proactive measure ahead of the pre-planned schedule: he briefly and proactively reduced the load on a non-critical load in a large industrial park. This operation, while not considered standard accident handling at the time, successfully created a stability margin for the power grid, effectively preventing the fault from escalating, buying valuable time for Dispatcher A's fault handling, and ultimately significantly shortening the time required for the entire network to return to stability.

[0063] In the intelligent analysis method for employee performance evaluation in this embodiment, the system first continuously acquires real-time change data of key power grid operating parameters, including voltage, frequency, power flow distribution, load status, and alarm information. This data is collected at millisecond or second-level frequencies via connection to SCADA / EMS and WAMS systems, and after cleaning and noise reduction, is stored in a time-series database. A data analysis module continuously monitors these parameters and calculates the real-time stability index of the power grid to reflect its current operating status.

[0064] Next, the system will identify non-standard operating procedures performed by dispatcher B. Dispatcher B's load reduction operation on the large industrial park, because it does not fall within the preset standard operating procedures, will be marked as a "non-standard operation" by the "operation comparison module." The system will immediately record detailed information about the operation, including dispatcher B's identity, operating instructions, and operation time, and extract snapshots of key power grid parameters before and after the operation from the time-series database, forming a "non-standard operation event package."

[0065] Then, for a grid fault scenario where dispatcher B performs non-standard operating procedures, the system generates an expected recovery path. A "scenario matching module" analyzes the type, severity, affected area, and initial grid operating state of the current fault, searching for the most similar historical events from the "past event experience summary library." The system then synthesizes the average recovery time, voltage / frequency recovery curve, and load recovery curve of the grid under strict adherence to standard operating procedures from these similar historical events to generate an "expected recovery path" as a benchmark.

[0066] Subsequently, the system tracks the actual changes in key power grid operating parameters after dispatcher B's non-standard operating behavior occurs, thus obtaining the actual recovery path. After the non-standard operation is identified, the system continuously retrieves real-time data of key power grid operating parameters from the time-series database and plots it as an "actual recovery path" curve.

[0067] Furthermore, the system compares the actual recovery path with the expected recovery path. A "path comparison module" compares the actual recovery path with the expected recovery path in real time, comparing indicators including recovery time, stability, and additional guarantees for grid stability. In this example, dispatcher B's load shedding operation successfully created a stability margin for the grid, effectively preventing the fault from escalating, significantly shortening the time for the entire network to return to stability, and avoiding potential secondary faults. Therefore, the actual recovery path is significantly better than the expected recovery path in terms of recovery time, stability, and additional guarantees. When the "path comparison module" finds that the actual recovery path is significantly better than the expected recovery path, the system activates the "contribution quantification module" to calculate the contribution score of dispatcher B's non-standard operating behavior to the improvement of grid stability and efficiency. For example, a positive contribution score is calculated based on preset weighting coefficients and additional guarantee coefficients.

[0068] Simultaneously, the system assesses the short-term disturbances caused by dispatcher B's non-standard operational behavior and their long-term positive impacts. A "property discrimination module" analyzes the fluctuations in grid parameters within a very short period after the load shedding operation, quantifying its "short-term disturbance value." Because the load shedding operation is brief and proactive, its short-term disturbance value is at a moderate level. However, this operation quickly leads to optimized grid power flow distribution, avoiding line overload tripping and bringing significant long-term positive impacts. Therefore, the system determines this operation as a "benevolent optimization."

[0069] Finally, based on the contribution score and evaluation results, the system will determine that dispatcher B's non-standard operating behavior is classified as "benevolent optimization," and calculate the corresponding employee points for dispatcher B accordingly. A "points summary module" will add the contribution score for this "benevolent optimization" to dispatcher B's regular points in other areas, and the final points will be updated in real time to dispatcher B's personal performance file. The system will also provide a detailed points report explaining how the contribution value was calculated, to enhance the transparency and persuasiveness of the evaluation results.

[0070] Through the above embodiments, the intelligent analysis method for employee performance evaluation in this application can effectively solve the problem that existing systems cannot accurately identify and reward the contribution of dispatchers' non-standard operating behaviors to the improvement of power grid stability and efficiency in power dispatch centers. Traditional existing performance evaluation systems, as described in the background section, when processing dispatcher B's preventative load reduction operation, because this operation does not fall within the scope of routine accident handling, the system's built-in anomaly detection mechanism marks it as abnormal behavior and assigns a lower evaluation value, even offsetting other positive points. This leads to inaccurate value judgments of dispatcher B, and fails to fairly measure the value of their expert-level decisions based on deep experience and prediction, which go beyond procedures.

[0071] In contrast, the method of this application achieves significant technical contributions through the following key technical points:

[0072] Firstly, regarding data acquisition, this application continuously collects real-time changes in key power grid operating parameters, and then cleans, denoises, and stores the data to ensure the precision and real-time nature of the analysis. This provides a solid data foundation for subsequent identification of non-standard operations and effect evaluation, avoiding misjudgments caused by insufficient or outdated data dimensions in traditional systems.

[0073] Secondly, regarding the identification of non-standard operations, this application establishes a "standard operating procedure library" and an "operation comparison module," enabling it to accurately identify non-standard operating behaviors performed by dispatchers and record a snapshot of the power grid state at the time of occurrence. This allows the system to capture expert-level decision-making behaviors that exceed standard operating procedures and are based on experience-based judgment and prediction, whereas traditional systems often simply label these behaviors as abnormal.

[0074] Furthermore, regarding effect evaluation, this application quantifies the contribution score of non-standard operations to grid stability and efficiency improvement by generating an "expected recovery path" and comparing it with the "actual recovery path." This dynamic comparison mechanism based on historical data and actual effects can objectively evaluate the actual value of non-standard operations, avoiding the limitations of traditional systems that simply judge based on whether they comply with procedures. For example, in the above example, although dispatcher B's load reduction operation was non-standard, the actual recovery path it brought was significantly better than expected, thus its contribution was accurately identified and quantified.

[0075] Furthermore, this application also obtained a comprehensive evaluation result by assessing the short-term disturbances caused by non-standard operating behaviors and their long-term positive impacts. This enables the system to comprehensively weigh the risks and benefits of operations, effectively distinguish between "benevolent optimization" and "misoperation" or "malicious sabotage," and avoid the problem of traditional systems focusing solely on short-term negative factors while ignoring long-term positive effects.

[0076] Finally, based on contribution scores and evaluation results, this application determines the operational nature of non-standard operating behaviors and calculates the corresponding dispatcher's employee points. This ensures that employees who make decisions that go beyond procedures but are beneficial to the overall power grid at critical moments receive fair recognition and incentives. This solves the problem that traditional systems cannot accurately identify and reward dispatchers' contributions to power grid stability and efficiency through non-standard operating behaviors, thus improving the fairness and accuracy of employee point evaluation.

[0077] In some embodiments, the specific steps in step S5 include:

[0078] S51. Obtain the real-time fault type, affected load, affected critical infrastructure, and deviation of key power grid operating parameters when non-standard operating behaviors occur;

[0079] S52. Assess the severity of the power grid based on the real-time fault type, the amount of load affected, the critical infrastructure affected, and the deviation of key power grid operating parameters; the severity of the fault reflects the risk level of cascading failures or system disconnection in the power grid.

[0080] S53. Adjust the weighting coefficients of the various assessment indicators used to calculate the contribution score according to the severity of the power grid crisis; the assessment indicators include recovery time, power grid stability, and additional power grid stability guarantees;

[0081] S54. Using the adjusted weighting coefficients, the differences between the actual recovery path and the expected recovery path in terms of evaluation indicators are weighted and calculated to obtain the contribution score of non-standard operating behavior to the improvement of power grid stability and efficiency.

[0082] In step S51, the real-time fault type refers to the specific fault mode faced by the power grid when non-standard operating behaviors occur, such as short-circuit faults, ground faults, open-circuit faults, and oscillation instability. These types can be identified through real-time analysis of relay protection action information, circuit breaker status changes, fault recording data, and power grid topology analysis. The affected load refers to the total load that is interrupted or restricted due to power grid faults or non-standard operating behaviors. This data can be obtained through a real-time load monitoring system and statistically analyzed in conjunction with user load data from the affected area. The affected critical infrastructure refers to important power equipment or facilities that are directly or indirectly affected by power grid faults, such as critical transmission lines, main transformers, important power plants, and large load centers. This information can be combined with geographic information systems (GIS) and power grid topology data to identify critical equipment near or electrically connected to the fault point. The deviation of key operating parameters of the power grid refers to the degree of deviation between the key operating parameters of the power grid, such as voltage, frequency, and power flow distribution, and the normal operating values ​​or preset safety thresholds when a fault occurs. These deviation data can be obtained by comparing the voltage, current, frequency, active power, reactive power, and other data collected in real time by the wide area measurement system (WAMS) or SCADA / EMS system with the normal operating reference values.

[0083] In step S52, assessing the severity of the power grid's urgency aims to quantify the severity and potential risks of the current power grid fault situation. This can be achieved by establishing a multi-factor comprehensive assessment model. This model takes the aforementioned data as input. For example, a fuzzy comprehensive evaluation method can be used to quantify and score factors such as real-time fault type, affected load, damage to critical infrastructure, and parameter deviations, and combine this with an expert experience rule base for comprehensive judgment. Alternatively, risk matrix analysis can be used to combine the probability of a fault occurrence with the severity of its consequences to derive the urgency level. The urgency level reflects the probability and severity of a cascading failure or system disconnection in the current state of the power grid. For example, the urgency level can be divided into three levels: low, medium, and high. "High urgency" may mean that the power grid is on the edge of its stability limit and is highly susceptible to large-scale power outages.

[0084] In step S53, adjusting the weighting coefficients aims to make the calculation of contribution scores more adaptable to the grid operation needs under different levels of urgency. This can be achieved through preset weighting adjustment rules or adaptive algorithms. For example, when the grid is in a "high-urgency" state, the system can significantly increase the weights of "grid stability" and "recovery time" to emphasize the importance of rapid recovery and avoiding secondary accidents; while in a "low-urgency" state, the weight of "additional grid stability protection" may be appropriately increased to encourage dispatchers to pursue more optimized operations while ensuring stability. Evaluation indicators are key dimensions for measuring the impact of non-standard operating behaviors on the grid. Among them, recovery time refers to the time required for the grid to recover from a fault state to a normal or acceptable operating state; grid stability refers to the fluctuation amplitude, oscillation suppression capability, and whether instability is avoided during the recovery process of the grid parameters such as voltage, frequency, and power flow; additional grid stability protection refers to the additional benefits provided by non-standard operations in the process of restoring grid stability, in addition to directly restoring stability, such as avoiding potential secondary faults, reducing unnecessary load reductions, and optimizing grid power flow distribution.

[0085] In step S54, weighted calculation is a method that comprehensively considers the impact of different evaluation indicators on the contribution score. It obtains a comprehensive contribution score by multiplying the difference of each evaluation indicator by its corresponding weight coefficient and then summing all weighted differences. This calculation method ensures that the contribution assessment of dispatchers' non-standard operations under different levels of urgency is more fair and accurate.

[0086] This application's solution introduces a dynamic assessment and weighting adjustment mechanism for the severity of grid urgency before calculating contribution scores, ensuring that the value judgment of dispatchers' non-standard operations is no longer static and one-size-fits-all. It is closely integrated with steps such as acquiring real-time grid data, identifying non-standard operations, generating expected recovery paths, tracking actual recovery paths, and comparing paths, forming a more comprehensive and intelligent evaluation system. Within this framework, this solution further refines the contribution score calculation process, ensuring that innovative operations by dispatchers that exceed procedures can receive a fairer and more accurate quantitative evaluation in complex and ever-changing grid fault scenarios. This dynamically adjusted evaluation method effectively solves the problem that traditional systems cannot accurately identify and measure the value of expert-level decisions based on deep experience and prediction in the face of complex emergency situations, thus avoiding the underestimation or overestimation of dispatchers' contributions.

[0087] The following is a concrete example. During a power grid fault, when a dispatcher performs a non-standard operation, the system immediately initiates the contribution score calculation process. First, the system interfaces with the SCADA / EMS and WAMS systems to obtain real-time power grid data at the time the non-standard operation occurred. For example, the system identifies the fault type as a "double-circuit near-zone short-circuit fault," affecting a load of 500MW, and impacting critical infrastructure including a 220kV substation and two important transmission lines. It also detects deviations in key operating parameters such as "multiple bus voltage drops exceeding 10%" and "a momentary drop in system frequency to 49.5Hz." Next, the system inputs this real-time data into a pre-set urgency assessment model. This model is based on a machine learning model trained using expert system rules and historical data. For example, based on the severity of the "double-circuit near-circuit fault," the significant load impact of "500MW," the criticality of the "220kV substation" and "important transmission lines," and the degree of "significant voltage and frequency deviation," the system assesses the current grid's urgency level as "high-critical," indicating a high risk of cascading failures and system disconnection. Subsequently, based on the "high-critical" assessment, the system retrieves or dynamically calculates a set of weighting coefficients from a pre-set weighting coefficient table. For instance, in a "high-critical" state, the system can set the weighting coefficient for "recovery time" to 0.5, the weighting coefficient for "grid stability" to 0.4, and the weighting coefficient for "additional grid stability protection" to 0.1. This reflects that rapid recovery and maintaining stability are the primary objectives in emergency situations. Finally, the system uses these adjusted weighting coefficients to perform a weighted calculation on the differences between the actual and expected recovery paths in terms of recovery time, grid stability, and additional grid stability protection. For example, if the actual recovery time is 10 minutes shorter than expected, the grid stability (such as voltage fluctuation) is 5% better than expected, and the non-standard operation avoids a potential secondary fault (corresponding to an additional protection factor of 1.5), the system will perform a weighted summation based on the adjusted weighting coefficients to calculate the final contribution score of the dispatcher's non-standard operation.

[0088] By employing the aforementioned technical solution, this application effectively addresses the problem that traditional employee performance evaluation systems, when assessing the contributions of power dispatchers' non-standard operations, fail to adequately consider the severity of grid urgency, leading to fixed evaluation weights and potentially underestimating or overestimating actual contributions. This solution acquires detailed data on grid fault scenarios in real time and dynamically assesses the severity of grid urgency accordingly, adjusting the weighting coefficients of various evaluation indicators to ensure more accurate and fair contribution score calculations. In high-risk scenarios, the system can prioritize contributions to recovery speed and grid stability, while in other scenarios, it can more comprehensively consider efficiency improvements and additional safeguards. This dynamic and adaptive evaluation mechanism ensures that dispatchers' innovative operations—which go beyond procedures and are based on experience and prediction—in the face of complex and ever-changing grid faults receive objective and fair quantitative recognition, effectively incentivizing dispatchers to make optimal decisions at critical moments and improving the overall safety and efficiency of grid operation.

[0089] In some embodiments, the specific steps in step S2 include:

[0090] S21. Identify non-standard operations performed by multiple dispatchers within a short period of time that are physically adjacent or electrically related in the power grid topology, and record the execution order and target of the non-standard operations;

[0091] The specific steps in step S6 include:

[0092] S61. For each identified non-standard operation, based on the grid state at the time of its occurrence, use the physical laws of the power system to predict the immediate and short-term impacts of the non-standard operation on key operating parameters such as grid voltage, frequency, and power flow distribution.

[0093] S62. Based on the execution sequence and electrical correlation of non-standard operations, the predicted immediate and short-term effects are physically superimposed or corrected to generate an operation interaction effect path that comprehensively reflects the interaction of all related non-standard operations.

[0094] S63. Track the actual changes in key power grid operating parameters after all associated non-standard operations occur to obtain the actual state evolution path;

[0095] S64. Compare the actual state evolution path with the operation interaction impact path to identify the deviation between the actual effect and the predicted interaction effect;

[0096] S65. When there is a significant deviation between the actual state evolution path and the operational interaction path, power system analysis techniques are used to reverse-analyze the physical causes of the deviation, decompose the independent contribution of each non-standard operation to the short-term disturbance and long-term positive impact on the power grid, and identify the physical enhancement, offsetting or secondary effects between non-standard operations.

[0097] S66. Based on the decomposed independent contributions, quantify the short-term disturbances caused by each non-standard operation and their long-term positive impacts to obtain the evaluation results.

[0098] In the above scheme, step S21 aims to identify non-standard operations that may affect each other. Here, "within a short time" typically refers to a time window of seconds to minutes to capture closely related operation sequences. "Physical proximity" refers to the geographical proximity of the operation objects, such as operations at the same substation or on adjacent lines; while "electrical correlation" refers to the electrical connection relationship of the operation objects in the power grid topology, such as equipment connected through the same bus or components on the same power flow path. Identifying these related operations can be done by combining spatial queries based on Geographic Information System (GIS) with time windows, or by analyzing the power grid topology map and real-time power flow data to determine the electrical connection path. For example, the system can maintain a power grid topology database and, combined with the device IDs in the operation logs, use graph algorithms (such as shortest path algorithms or connected component algorithms) to determine the electrical correlation between operation objects.

[0099] Step S61 is used to predict the impact of a single non-standard operation on the power grid. The "physical laws of the power system" form the basis for this prediction, including but not limited to Kirchhoff's laws, Ohm's law, power balance equations, and transient stability equations. These laws constitute the mathematical model of power system operation. Predicting "immediate and short-term impacts" can be achieved using various power system analysis tools. For example, steady-state and transient simulations can be performed using power flow calculation software (such as PSCAD / EMTDC, PSS / E) to assess the impact of the operation on voltage, frequency, active / reactive power flow, and power angle stability. The prediction results can be represented as curves or values ​​showing the changes in key power grid operating parameters over a period of time after the operation.

[0100] Step S62 is crucial for handling the interactions of multiple non-standard operations. "Physically superimposing or correcting" the predicted immediate and short-term effects means more than simply adding the effects of individual operations; it involves considering the sequence of operations and how they mutually alter the grid state, thus affecting the effectiveness of subsequent operations. For example, a dispatcher's load shedding operation might change the grid's voltage level, thereby affecting the effectiveness of another dispatcher's reactive power compensation equipment switching operation. This can be achieved through sequential simulation or iterative calculation: after simulating the effect of the first operation, the grid state is updated, and the effect of the second operation is simulated based on this updated state, and so on, thus generating a comprehensive "operational interaction path."

[0101] Step S63 obtains the "actual state evolution path" by continuously monitoring key operating parameters of the power grid. This is similar to obtaining real-time change data in step S1, but focuses more on the continuous tracking of parameters such as grid voltage, frequency, power flow distribution, and equipment status within a specific time period after non-standard operations occur. This data typically comes from power dispatch automation systems (SCADA / EMS) and wide-area measurement systems (WAMS) and is stored in a high-performance time-series database.

[0102] Step S64 identifies the "deviation" between the "actual state evolution path" and the "operational interaction impact path" by comparing the two. This comparison can employ various quantification methods, such as calculating the root mean square error (RMSE) of the two paths on key parameters, the maximum deviation value, or using the Dynamic Time Warping (DTW) algorithm to measure the similarity of the curves. When the deviation exceeds a preset threshold, it is considered to have a "significant deviation," indicating that the predictive model has failed to fully capture the actual grid response or the complex interactions between operations.

[0103] Step S65, upon identifying a significant deviation, utilizes "power system analysis techniques" for in-depth reverse analysis. This includes, but is not limited to, sensitivity analysis, fault source analysis, state estimation, or artificial intelligence-based pattern recognition techniques. These techniques can trace the physical causes of the deviation; for example, the actual value of a device parameter may not match the model, or there may be external disturbances not considered by the model. Simultaneously, the core of this step lies in "decomposing the independent contribution of each of the aforementioned non-standard operations to the short-term disturbances and long-term positive impacts on the power grid." This can be achieved by constructing multivariate regression models, causal inference models, or game theory-based methods to quantify the true role of each operation in a complex interactive environment. Furthermore, identifying "physical enhancement, offsetting, or secondary effects" requires analyzing whether, in the operation sequence, one operation amplifies the positive or negative effects of another operation (enhancement), weakens the effect of another operation (offset), or triggers new, unexpected power grid phenomena (secondary effects).

[0104] Step S66 quantifies the short-term disturbances and long-term positive impacts caused by each non-standard operation based on the independent contributions decomposed in S65, thereby obtaining the final evaluation result. This can be achieved using a weighted scoring model, assigning weights to short-term disturbances (such as voltage drop magnitude and peak frequency deviation) and long-term positive impacts (such as shortened recovery time and improved grid stability), and calculating based on the decomposed independent contributions.

[0105] The proposed solution, through the aforementioned steps, effectively addresses the challenge of accurately assessing the short-term disturbances and long-term positive impacts of multiple dispatchers performing interconnected non-standard operations during complex power grid fault handling in power dispatch control centers. First, S21 identifies groups of non-standard operations with physical proximity or electrical correlation, laying the foundation for subsequent interaction analysis. Next, S61 and S62 utilize the physical laws of power systems to predict the impact of individual operations, considering the execution sequence and electrical correlation to generate a comprehensive "operation interaction impact path" reflecting the interaction, making predictions for complex scenarios more realistic. Subsequently, S63 tracks the actual power grid state, and S64 compares the actual state with the predicted path to identify deviations. When significant deviations exist, S65 utilizes advanced power system analysis techniques to analyze the physical causes of the deviations in reverse, accurately "decomposing the independent contribution of each non-standard operation to the short-term disturbances and long-term positive impacts of the power grid," while identifying physical enhancements, offsets, or secondary effects between operations. This significantly improves the accuracy and fairness of the assessment. Ultimately, S66 quantifies the short-term disturbances and long-term positive impacts of each operation based on the decomposed independent contributions, providing a reliable basis for subsequent employee performance evaluation. This method avoids the evaluation distortion caused by simply treating multiple operations as independent events, enabling a fairer and more objective evaluation of expert-level decisions made by dispatchers that exceed procedures in complex emergency situations. Combined with basic intelligent analysis methods for employee performance evaluation, this solution ensures more accurate calculation of contribution scores and evaluation results for non-standard operations in complex scenarios involving multiple dispatchers and multiple operations. This allows the entire performance evaluation system to more accurately identify and incentivize dispatchers who truly contribute to improving grid stability and efficiency.

[0106] The following is a concrete example to illustrate this. Suppose that during a power grid failure caused by severe regional convective weather, multiple transmission lines tripped consecutively, resulting in a significant voltage drop and frequency fluctuation in the local power grid. Dispatcher B, without clear procedural support, implemented preventative load shedding (non-standard operation A) on a large industrial park to stabilize the regional voltage. Almost simultaneously, dispatcher C in a neighboring area also performed a non-standard operation, switching on a set of reactive power compensation devices (non-standard operation B) to improve the voltage in his area.

[0107] First, the system identifies dispatcher B's load reduction operation A and dispatcher C's reactive power compensation switching operation B in S21. Because they occur within a short time and are electrically related in the grid topology (e.g., affecting each other through the same substation or adjacent lines), they are identified as a set of related non-standard operations, and their execution sequence is recorded. Next, in S61, based on the grid state at the time of operation A and operation B, the system uses power flow calculations and transient stability simulations to predict the immediate and short-term effects of operation A on grid voltage, frequency, and power flow distribution, and the immediate and short-term effects of operation B on grid voltage and reactive power flow, respectively. In S62, the system physically superimposes and corrects these predicted effects according to the execution sequence and electrical correlation of operation A and operation B. For example, the load reduction operation A will first change the power flow distribution and voltage level of the grid, and the reactive power compensation switching operation B will further affect the voltage based on this changed grid state. The system generates an "operation interaction effect path" that comprehensively reflects the interaction between A and B.

[0108] Subsequently, in S63, the system continuously tracks the actual changes in key power grid operating parameters after operations A and B occur, obtaining the "actual state evolution path." In S64, the system compares this actual state evolution path with the previously generated predicted interaction path to identify the deviation between the two. If the system finds that the actual voltage recovery speed is faster than predicted and the frequency fluctuation is smaller, it indicates a significant deviation.

[0109] At this point, in S65, the system utilizes power system analysis techniques, such as sensitivity analysis and fault tracing, to reverse-analyze the physical causes of this deviation. The system may discover that dispatcher B's load shedding operation A not only directly stabilized the voltage but also indirectly reduced line losses, allowing dispatcher C's reactive power compensation switching operation B to play a greater role under more optimized grid conditions, thus producing a "physical enhancement" effect. Through this analysis, the system can decompose the independent contribution of operation A to grid stability, as well as the independent contribution of operation B under the influence of operation A. For example, operation A independently contributed X% improvement in voltage drop amplitude, while operation B contributed Y% improvement on top of that, and the combined effect of both shortened the recovery time by Z minutes. Finally, in S66, based on these decomposed independent contributions, the system quantifies the short-term disturbances (e.g., load shedding amount, instantaneous voltage fluctuations) and long-term positive effects (e.g., avoiding cascading failures, shortening recovery time) caused by dispatcher B's operation A and dispatcher C's operation B, respectively, obtaining their respective evaluation results.

[0110] Through the above technical solution, this application can effectively identify and handle the complex interactions between multiple non-standard operations, avoiding evaluation distortion caused by simple superposition or ignoring interactions. This allows for a more accurate decomposition and quantification of the independent contributions of each dispatcher in complex power grid fault scenarios, especially when multiple dispatchers simultaneously perform non-standard operations. Therefore, this solution ensures a fairer and more objective employee performance evaluation for dispatchers, incentivizing them to make decisions that go beyond procedures but are beneficial to the overall power grid at critical moments based on experience and foresight, thereby improving the overall stability and efficiency of power grid operation.

[0111] In some embodiments, the steps of using power system analysis techniques to reverse-engineer the physical causes of the deviations and decompose the independent contribution of each non-standard operation to the short-term disturbances and long-term positive effects on the power grid include:

[0112] S65A1. Acquires power grid topology and real-time operational data; real-time operational data includes key power grid operating parameters;

[0113] S65A2. Based on the power grid topology and real-time operation data, and combined with the preset electrical characteristic association rules of power grid components, the physical propagation path of the deviation is obtained by tracing the deviation between the actual state evolution path and the operation interaction influence path step by step.

[0114] S65A3. Identify the key physical nodes or branches that cause deviations along the physical propagation path;

[0115] S65A4. Based on the degree of impact of non-standard operations on critical physical nodes or branches, decompose the independent contribution of each non-standard operation to the short-term disturbances and long-term positive impacts on the power grid.

[0116] In some implementations of the above schemes, acquiring the power grid topology and real-time operational data is fundamental for subsequent analysis. Power grid topology refers to the connection relationships and physical layout of components such as generators, transformers, transmission lines, buses, and loads within the power grid. This topology can be obtained from a Geographic Information System (GIS) or Network Model Management System (NMMS), or generated in real-time through the topology analysis function of a Supervisory Control and Automation System (SCADA / EMS). Real-time operational data reflects the operating status of the power grid at a specific moment, including but not limited to key operating parameters such as grid voltage, frequency, power flow distribution, active power, reactive power, circuit breaker status, and relay protection operation information. This data can be collected in real-time from SCADA / EMS systems, Wide Area Measurement Systems (WAMS), or Phasor Measurement Units (PMUs) and stored in high-performance time-series databases, such as InfluxDB or TimescaleDB.

[0117] Pre-defined rules governing the electrical characteristics of power grid components are crucial for guiding deviation tracing. These rules can be based on the physical laws of power systems, such as Ohm's law, Kirchhoff's laws, and power flow equations, or on power grid operation procedures and relay protection coordination logic. These rules define the electrical interactions between power grid components and how changes in the power grid state propagate through these components. Step-by-step tracing refers to starting from the deviation between the observed actual evolution path of the power grid state and the operational interaction path, and using these pre-defined rules, gradually deduce the physical root cause and propagation path of the deviation. This can be done through simulation analysis based on physical models, expert system reasoning, or by combining machine learning algorithms. The final physical propagation path of the deviation is the actual physical link in the power grid from the source of the deviation to the observation point.

[0118] Identifying the critical physical nodes or branches causing the deviation refers to locating the power grid components that have the most significant impact on the deviation along a determined physical propagation path. These critical physical nodes or branches can be specific substation buses, transmission lines, transformers, or generator units. Identification methods can include sensitivity analysis, which assesses the degree of impact of changes in different component parameters on the overall state of the power grid; or it can utilize fault location algorithms, combined with real-time measurement data and power grid models, to accurately calculate the location of the fault point.

[0119] Decomposing the independent contribution of each non-standard operation to the short-term disturbances and long-term positive impacts on the power grid is the core of quantifying the value of non-standard operations. The degree of impact can be quantified as the magnitude or duration of changes in key operating parameters of the power grid, such as voltage, frequency, power flow distribution, and load, caused by the non-standard operation. The decomposition of independent contributions can be achieved in various ways. For example, using power system simulation tools for "what-if analysis," that is, simulating the grid response with and without a specific non-standard operation, to calculate the independent impact of the operation; or using the Shapley value method based on cooperative game theory, or using machine learning models for feature contribution analysis, to objectively quantify the independent value of each non-standard operation to the short-term disturbances and long-term positive impacts on the power grid.

[0120] This application's solution systematically acquires power grid topology and real-time operational data, ensuring that the analysis is based on the actual physical structure and real-time operating status of the power grid, avoiding evaluation biases caused by incomplete data. Based on this, and combined with pre-defined electrical characteristic association rules for power grid components, the deviation between the actual state evolution path and the operational interaction path is traced step-by-step to systematically track the physical root causes of the deviations, preventing blind analysis and ensuring that the physical propagation path truly reflects the chain effects of the power grid. Furthermore, identifying key physical nodes or branches causing deviations along the physical propagation path allows focusing on core problem points in the power grid, improving analysis efficiency and directly locating the source of influence. Finally, the independent contribution of each non-standard operation is decomposed based on its impact on key physical nodes or branches, and the impact of the operation on specific nodes is quantified, ensuring the objectivity and fairness of the decomposition, avoiding subjective assumptions, and thus accurately distinguishing the independent value of each operation for short-term disturbances and long-term positive impacts. This scheme, combined with the aforementioned technical means of identifying non-standard operations performed by multiple dispatchers within a short period of time that have physical proximity or electrical correlation in the power grid topology and recording the execution order and objects of the non-standard operations, can more accurately analyze the interaction between multiple non-standard operations. On this basis, it can quantify the independent contribution of each non-standard operation in detail, thereby providing a more solid and objective basis for subsequent evaluation and integral calculation.

[0121] The following is a concrete example to illustrate this. Suppose that during a power grid fault, dispatcher A and dispatcher B each performed two non-standard operations with electrical correlation. The system first obtains the latest power grid topology from the power grid GIS system and collects key operating parameters such as voltage, frequency, and power flow in real time from the SCADA / EMS and PMU systems during the fault. When the system identifies a significant deviation between the actual state evolution path and the operational interaction path, it initiates reverse analysis. The system uses the power flow calculation model and relay protection action logic as preset electrical characteristic correlation rules. Starting from observed deviations such as a persistently low voltage on a bus or abnormal power flow fluctuations on a line, it uses iterative calculations or graph search algorithms to progressively deduce the physical propagation path leading to these deviations. For example, it traces back from the affected load point to the power supply substation, then to the upstream transmission line, until it finds the initial fault point or operational impact point. Along this physical propagation path, the system identifies key physical nodes, such as a transmission line with excessive power flow shift due to the operation, or a substation bus with insufficient voltage support due to the operation. Subsequently, the system performs independent simulation analyses for each non-standard operation, such as dispatcher A's reactive power compensation device switching operation and dispatcher B's load reduction operation. By comparing the grid response in three scenarios—operating only dispatcher A's operation, operating only dispatcher B's operation, and operating both operations simultaneously—the system quantifies the impact of each operation on parameters such as voltage and power flow at key physical nodes. For example, it calculates how much dispatcher A's operation reduced the power flow of the transmission line and how much dispatcher B's operation increased the voltage of the bus. Based on these quantification results, the system can decompose the independent contribution of each non-standard operation to short-term grid disturbances (such as voltage fluctuations at the moment of operation) and long-term positive effects (such as avoiding line overloads or improving system voltage stability).

[0122] Through the above technical solution, this application can systematically acquire basic power grid data, achieve physical tracing of deviation paths, focus on key physical nodes or branches causing deviations, and quantify and decompose the impact of non-standard operations on these nodes or branches. This effectively solves the problems of incomplete data, ambiguous tracing, and strong subjectivity in decomposition in traditional evaluation methods, thereby ensuring that the evaluation results of non-standard operational behaviors are more accurate, objective, and fair. Especially in complex scenarios where multiple non-standard operations interact, this solution can accurately distinguish the independent value of each operation, providing solid technical support for identifying and incentivizing dispatchers who make decisions that go beyond procedures but are beneficial to the overall power grid at critical moments.

[0123] In some embodiments, the step of identifying physical enhancements, cancellations, or secondary effects between non-standard operations includes:

[0124] S65B1. Collect time-series data of key power grid operating parameters before and after non-standard operations occur, as well as dispatcher operation logs;

[0125] S65B2. Using event sequence analysis technology, combined with time series data of key power grid operating parameters and dispatcher operation logs, identify nonlinear correlation patterns between power grid state changes and multiple non-standard operations within a specific time window;

[0126] S65B3. Based on nonlinear correlation patterns, it identifies physical enhancements, cancellations, or secondary effects between nonstandard operations.

[0127] To implement the above technical solutions, it is necessary to collect time-series data of key power grid operating parameters before and after non-standard operations, as well as dispatcher operation logs. This step aims to provide a comprehensive and real-time data foundation for subsequent analysis, ensuring that the analysis process is based on the actual operating state of the power grid rather than theoretical assumptions. Specifically, this can be achieved by establishing data interfaces with the power dispatch automation system (SCADA / EMS) and the wide-area measurement system (WAMS) to acquire key operating parameters such as voltage, current, frequency, active power, reactive power, and circuit breaker status of various nodes in the power grid (e.g., substations, power plants) in real time, and storing them at high frequency (e.g., milliseconds or seconds) in a high-performance time-series database. Simultaneously, the dispatcher operation log system can record all dispatcher operation instructions in real time, including operator identity, operation instructions, operation time, and operation object, and store them in a structured database or log management system. Furthermore, a data integration platform can be used to uniformly collect, clean, and standardize data from different sources (e.g., SCADA, WAMS, operation log systems), and align them according to timestamps to form a complete dataset.

[0128] Building upon this foundation, event sequence analysis techniques are employed, combining time-series data of key power grid operating parameters with dispatcher operation logs, to identify nonlinear correlation patterns between power grid state changes and multiple non-standard operations within a specific time window. The core of this step lies in capturing the complex causal relationships between operation sequences and power grid dynamics, identifying nonlinear correlation patterns that are difficult to detect using traditional linear methods. Specifically, rule-based event sequence matching algorithms can be used, predefining a series of possible operation combinations and their corresponding power grid state change patterns, and identifying them through pattern matching. Alternatively, machine learning methods, such as Hidden Markov Models (HMMs), Recurrent Neural Networks (RNNs), or Long Short-Term Memory Networks (LSTMs), can be used to train time-series data and operation logs to learn and identify nonlinear correlation patterns between power grid state changes and non-standard operations. Furthermore, statistical methods such as Granger causality tests or mutual information can be used to analyze the lag correlations and nonlinear dependencies between different operation sequences and changes in power grid parameters.

[0129] Finally, based on the nonlinear correlation patterns, the physical enhancement, cancellation, or secondary effects between the nonstandard operations are identified. This step aims to objectively distinguish the cooperative or conflicting relationships between operations based on the identified nonlinear correlation patterns, thereby accurately quantifying the interaction contributions. Specifically, based on a preset expert rule base, the identified nonlinear correlation patterns can be matched with known rules for physical enhancement, cancellation, or secondary effects to determine the specific effect type. For example, if two operations cause a voltage drop in the same area within a short period of time that is less than the sum of the effects of a single operation, it may be a cancellation effect. Alternatively, the identified nonlinear correlation patterns can be input into a power grid simulation model. By simulating the impact of different combinations of operations on the power grid and comparing it with the impact of a single operation, the physical enhancement, cancellation, or secondary effects can be quantified and identified.

[0130] This application's solution, by establishing a comprehensive data foundation and utilizing advanced event sequence analysis technology, can deeply uncover complex nonlinear correlation patterns between multiple nonstandard operations and changes in grid state. First, by collecting time-series data of key grid operating parameters before and after nonstandard operations, as well as dispatcher operation logs, a solid data foundation is provided for subsequent analysis, ensuring its accuracy and reliability. Second, using event sequence analysis technology, nonlinear correlation patterns between grid state changes and multiple nonstandard operations within a specific time window can be effectively identified. This allows the system to capture complex interactions that are difficult to detect using traditional methods. Finally, based on these identified nonlinear correlation patterns, the system can accurately identify physical enhancements, cancellations, or secondary effects between nonstandard operations. This structured identification method enables a more accurate decomposition of the independent contribution of each operation to the grid when evaluating dispatcher nonstandard operations, especially in complex scenarios where multiple dispatchers perform nonstandard operations simultaneously, effectively distinguishing between cooperative or conflicting relationships between operations. This significantly improves the granularity and accuracy of nonstandard operation value assessment, thus providing a fairer and more reliable basis for employee performance evaluation.

[0131] The following is a concrete example to illustrate this. Suppose that during a power grid fault, dispatcher A performs a non-standard operation A (e.g., prematurely disconnecting a line before fully confirming the nature of the fault), while dispatcher B performs a non-standard operation B (e.g., proactively adjusting the output of a generator unit when the power flow is abnormal). To identify the physical effects between these two non-standard operations, the system first collects millisecond-level time-series data of key operating parameters in the power grid (e.g., voltage, frequency, power flow distribution) before and after operations A and B, as well as the operation logs of dispatchers A and B. Subsequently, the system uses a deep learning-based event sequence analysis model to analyze this data. This model, after training, can identify the non-linear impact patterns of operations A and B on power grid voltage stability within a specific time window (e.g., operation B immediately follows operation A within 30 seconds). For example, the model might identify that when operations A and B are performed in a specific order and at specific time intervals, the power grid voltage recovers to a stable state faster and with smaller voltage fluctuations than when either operation is performed alone or their effects are simply superimposed. Based on this identified nonlinear correlation pattern, the system can determine that there is a physical enhancement effect between operation A and operation B, meaning that they work synergistically to improve the stability of the power grid. Conversely, if the model identifies that the simultaneous execution of both operations leads to unexpected frequency oscillations, it may be identified as a secondary effect.

[0132] Through the above technical solution, this application effectively solves the problem that traditional methods, when evaluating non-standard operations, suffer from inaccurate evaluation results and an inability to accurately capture the complex interactions between operations due to the lack of specific data collection and analysis mechanisms. By comprehensively collecting time-series data of key power grid operating parameters and dispatcher operation logs before and after non-standard operations, a solid data foundation is provided for analysis. Utilizing event sequence analysis technology to identify nonlinear correlation patterns enables the system to capture more complex and detailed causal relationships between operation sequences and power grid dynamics. Ultimately, based on these nonlinear correlation patterns, the physical enhancement, cancellation, or secondary effects between non-standard operations can be accurately identified, thereby significantly improving the precision and accuracy of non-standard operation value assessment. This allows for a more fair and reliable evaluation of dispatchers' non-standard operational behaviors when facing sudden, complex, and high-risk power grid fault scenarios in power dispatch control centers, thereby improving the fairness and reliability of employee performance evaluation.

[0133] In some embodiments, step S7, which involves determining the operational nature of non-standard operational behavior based on contribution scores and evaluation results, includes:

[0134] S71. Set their respective weighting coefficients based on the relative importance of contribution scores, short-term disturbances caused by non-standard operating behaviors, and their long-term positive impacts.

[0135] S72. Based on the weighting coefficients, a weighted comprehensive calculation is performed on the contribution score, the short-term disturbances caused by non-standard operating behavior, and the long-term positive impacts they bring, to obtain a comprehensive evaluation index for non-standard operating behavior.

[0136] S73. Determine the operational nature of non-standard operating behaviors based on the numerical range of the comprehensive evaluation indicators.

[0137] This scheme aims to quantify the relative importance of different evaluation dimensions in judging the nature of non-standard operating behaviors. By setting weighting coefficients, the system can flexibly reflect the degree of influence of contribution scores, short-term disturbances, and long-term positive impacts on the final evaluation results under different power grid operating scenarios. One implementation method is for an expert group composed of senior power dispatching experts or managers to assess the importance of each indicator and manually set weighting coefficients based on their rich experience and deep understanding of power grid operation risks. These weighting coefficients can be periodically adjusted according to the seasonal characteristics of power grid operation, load characteristics, or specific events (such as major power supply guarantee tasks). Another implementation method is for the system to utilize historical data and machine learning algorithms, such as regression analysis or decision tree models, to learn and automatically adjust the weighting coefficients. For example, by analyzing the actual effects of non-standard operations and expert evaluation results in a large number of past fault events, the model can be trained to identify which indicators should be assigned higher weights under what power grid conditions, thereby achieving adaptive adjustment of weights.

[0138] Based on this, the core of this scheme lies in integrating the evaluation results from multiple dimensions into a unified quantitative indicator to provide a comprehensive and objective evaluation of non-standard operating behaviors. Through weighted comprehensive calculation, the one-sidedness of a single indicator can be avoided, and the overall value of the operation can be reflected more accurately. One implementation method is to use a linear weighted summation method. For example, the comprehensive evaluation indicator can be expressed as: Comprehensive Evaluation Indicator = W_Contribution * Contribution Score + W_Disturbance * Short-Term Disturbance Value + W_Impact * Long-Term Positive Impact Value, where W_Contribution, W_Disturbance, and W_Impact are the weight coefficients set in step S71. The short-term disturbance value is usually represented by a negative value or a negative indicator to reflect its negative effect. Another implementation method is to use a multi-level fuzzy comprehensive evaluation method. This method allows the combination of qualitative factors (such as expert experience) with quantitative data. By constructing a fuzzy relation matrix and weight vector, fuzzy comprehensive calculations are performed on each indicator to obtain a more robust comprehensive evaluation indicator, which is particularly suitable for situations where there is a certain degree of fuzziness and uncertainty among the indicators.

[0139] Ultimately, this scheme aims to transform quantified comprehensive evaluation indicators into practically meaningful operational classifications, providing a basis for subsequent employee point calculations. By setting clear numerical ranges, the judgment of operational characteristics can be automated and standardized. One approach is to pre-define a series of discrete numerical ranges, each corresponding to a specific operational characteristic. For example, when the comprehensive evaluation indicator is greater than a certain threshold A, it is judged as "benevolent optimization"; when it is between threshold B and threshold A, it is judged as "effective operation"; when it is between threshold C and threshold B, it is judged as "neutral operation"; and when it is less than threshold C, it is judged as "erroneous operation" or "negative operation". These thresholds can be set based on historical data analysis and expert experience. Another approach is to use classification algorithms, such as Support Vector Machines (SVM) or neural networks, to classify the comprehensive evaluation indicators. By training the model, it can automatically classify non-standard operational behaviors into predefined operational characteristics, such as "benevolent optimization", "erroneous operation", or "malicious sabotage", based on the numerical value of the comprehensive evaluation indicator. This method can handle more complex nonlinear classification boundaries and improve the accuracy of judgments.

[0140] This application's solution addresses potential misjudgments in assessing the nature of non-standard operational behaviors by introducing weight settings, weighted calculations, and numerical range judgments. Specifically, in step S71, the system sets weight coefficients for the contribution score, the short-term disturbance caused by the non-standard operational behavior, and the relative importance of its long-term positive impact. This mechanism allows the system to dynamically adjust the weights of each indicator to reflect the differences in the importance of various factors under different power grid operating scenarios, thereby avoiding evaluation bias caused by fixed weights. For example, in extreme emergency situations where the power grid faces the risk of collapse, the weight of long-term positive impacts (such as avoiding system disconnection) may be significantly higher than that of short-term disturbances, while under normal operating conditions, the weight of short-term disturbances may be relatively higher. Based on this, step S72 performs a weighted comprehensive calculation on the contribution score, short-term disturbances, and long-term positive impacts based on these set weight coefficients, thereby obtaining a comprehensive evaluation index that can fully reflect the overall value of non-standard operational behaviors. This comprehensive calculation integrates multiple evaluation dimensions, solving the problem that a single indicator cannot fully reflect the operational effect, and ensuring the comprehensiveness and objectivity of the evaluation. Finally, step S73 determines the operational nature of non-standard operating behaviors based on the numerical range of the comprehensive evaluation index. This provides an objective and quantitative standard, ensuring that the judgment of the operational nature is based on comprehensive scores rather than subjective assumptions, thereby improving the accuracy and fairness of decision-making. This scheme is closely integrated with the basic method to form a more complete evaluation system. The basic method is responsible for acquiring key power grid operating parameters, identifying non-standard operating behaviors, generating expected recovery paths, tracking actual recovery paths, and calculating the contribution scores of non-standard operating behaviors to the improvement of power grid stability and efficiency, as well as assessing the short-term disturbances and long-term positive impacts they cause. This scheme, on the other hand, further provides a refined and intelligent mechanism to deeply integrate and analyze these preliminary quantitative results (contribution scores and evaluation results). By introducing weighting coefficients, this scheme can flexibly adjust the relative importance of various indicators according to the actual power grid operation context and management objectives, thereby more accurately determining the true nature of non-standard operating behaviors. This combination enables the system not only to identify the objective impact of non-standard operations, but also to make value judgments on them, effectively distinguishing between innovative operations that go beyond the rules but are beneficial to the overall power grid and erroneous operations that may have negative consequences. This provides a more fair and scientific basis for the evaluation of dispatchers' employee performance.

[0141] The following is a concrete example. Suppose that during a power grid fault, the dispatcher performs a non-standard operation. The system first calculates the contribution score of this operation (e.g., a positive score due to shortened recovery time) based on a basic method, and assesses its short-term disturbance (e.g., a small local voltage fluctuation caused by the operation) and its long-term positive impact (e.g., preventing the fault from escalating). As a specific implementation, in step S71, the system can dynamically set weighting coefficients based on the current operating state of the power grid (e.g., whether it is in an emergency, fault level, etc.). For example, if the power grid is in a high emergency state, the system might set the weighting coefficient for long-term positive impact to 0.6, the weighting coefficient for contribution score to 0.3, and the weighting coefficient for short-term disturbance to 0.1 to emphasize the importance of avoiding greater risks. These weighting coefficients can be stored in a configurable rule base and reviewed and adjusted by power grid operation and management personnel. In step S72, the system uses these weighting coefficients to perform a weighted comprehensive calculation of various indicators. For example, if the contribution score is +10 points, the short-term disturbance score is -2 points (negative values ​​indicate negative impact), and the long-term positive impact score is +15 points, then the comprehensive evaluation index can be calculated as: Comprehensive evaluation index = 0.3*(+10) + 0.1*(-2) + 0.6*(+15) = 3 - 0.2 + 9 = 11.8. In step S73, the system will determine the operational nature of the non-standard operation based on the value of this comprehensive evaluation index. For example, the system can preset the following judgment rules: if the comprehensive evaluation index is greater than 10, it is judged as "benevolent optimization"; if the comprehensive evaluation index is between 5 and 10, it is judged as "effective operation"; if the comprehensive evaluation index is between 0 and 5, it is judged as "neutral operation"; if the comprehensive evaluation index is less than 0, it is judged as "erroneous operation". Based on the above calculation result of 11.8, the non-standard operation will be judged as "benevolent optimization". In this way, the system can quantify and identify the value of expert-level decisions that go beyond procedures but are beneficial to the overall power grid in complex emergency situations, thus providing a fair basis for dispatchers' employee performance evaluation.

[0142] Through the aforementioned technical solution, this application can more accurately identify and quantify non-standard operating behaviors that exceed standard operating procedures but have a positive impact on grid stability and efficiency, effectively distinguishing between "well-intentioned optimization" and "mistaken operation." This solves the problem that traditional point-based systems cannot accurately measure the value of dispatchers' decisions based on deep experience and foresight in complex and emergency situations. This makes the employee point evaluation process more fair and transparent, truly incentivizing dispatchers to leverage their professional judgment and innovation capabilities at critical moments, thereby improving the overall safety and reliability of grid operation.

[0143] Reference Appendix Figure 2This invention provides an intelligent analysis system for employee point evaluation (this intelligent analysis system for employee point evaluation adopts the intelligent analysis method for employee point evaluation described in the above embodiments, and the specific process is described in the corresponding steps above), including:

[0144] The acquisition module 100 is used to acquire real-time change data of key operating parameters of the power grid;

[0145] The identification module 200 is used to identify non-standard operating behaviors performed by the dispatcher; non-standard operating behaviors are instructions or adjustments that do not fall within the preset standard operating procedures.

[0146] The generation module 300 is used to generate an expected recovery path for power grid fault scenarios when non-standard operating behaviors occur, by referring to the power grid recovery performance under standard operating procedures in historical fault events.

[0147] The monitoring module 400 is used to track the actual changes in key power grid operating parameters after non-standard operating behaviors occur, and to obtain the actual recovery path.

[0148] The comparison module 500 is used to compare the actual recovery path with the expected recovery path, and when it is determined that the actual recovery path is better than the expected recovery path, calculate the contribution score of non-standard operating behavior to the improvement of grid stability and efficiency.

[0149] The assessment module 600 is used to assess the short-term disturbances caused by non-standard operating behaviors and their long-term positive effects, and to obtain the assessment results.

[0150] The judgment module 700 is used to determine the nature of non-standard operating behavior based on contribution score and evaluation results, and to calculate the corresponding dispatcher's employee points based on the determined nature of the operation.

[0151] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0152] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent analysis method for employee score selection, characterized in that, Includes the following steps: S1. Obtain real-time change data of key power grid operating parameters; S2. Identify non-standard operating behaviors performed by the dispatcher; the non-standard operating behaviors are instructions or adjustments that do not fall within the preset standard operating procedures. S3. For the power grid fault scenario when the non-standard operating behavior occurs, refer to the power grid recovery performance under standard operating procedures in historical fault events to generate an expected recovery path; S4. Track the actual changes in key power grid operating parameters after the occurrence of the non-standard operating behavior to obtain the actual recovery path; S5. Compare the actual recovery path with the expected recovery path, and when it is determined that the actual recovery path is better than the expected recovery path, calculate the contribution score of the non-standard operating behavior to the improvement of power grid stability and efficiency; S6. Evaluate the short-term disturbances caused by the aforementioned non-standard operating behaviors and their long-term positive effects, and obtain the evaluation results; S7. Based on the contribution score and the evaluation result, determine the operational nature of the non-standard operation behavior, and calculate the corresponding dispatcher's employee points based on the determined operational nature.

2. The employee point selection intelligent analysis method of claim 1, wherein, The specific steps in step S5 include: S51. Obtain the real-time fault type, affected load, affected critical infrastructure, and deviation of key power grid operating parameters when the non-standard operating behavior occurs; S52. Assess the severity of the power grid based on the real-time fault type, the amount of load affected, the critical infrastructure affected, and the deviation of key power grid operating parameters; S53. Adjust the weighting coefficients of the various evaluation indicators used to calculate the contribution score according to the severity of the power grid crisis; S54. Using the adjusted weighting coefficients, the difference between the actual recovery path and the expected recovery path on the evaluation index is weighted and calculated to obtain the contribution score of the non-standard operating behavior to the improvement of power grid stability and efficiency.

3. The employee point selection intelligent analysis method of claim 2, wherein, The evaluation metrics include recovery time, grid stability, and additional grid stability guarantees.

4. The employee point selection intelligent analysis method of claim 1, wherein, The specific steps in step S2 include: S21. Identify non-standard operations performed by multiple dispatchers within a short period of time that have physical proximity or electrical correlation in the power grid topology, and record the execution order and operation objects of the non-standard operations; The specific steps in step S6 include: S61. For each identified non-standard operation, based on the power grid state at the time of its occurrence, and using the physical laws of the power system, predict the immediate and short-term impact of the non-standard operation on key operating parameters. S62. Based on the execution order and electrical correlation of the non-standard operations, the predicted immediate and short-term effects are physically superimposed or corrected to generate an operational interaction influence path that comprehensively reflects the interaction of all related non-standard operations. S63. Track the actual changes in key power grid operating parameters after all associated non-standard operations occur to obtain the actual state evolution path; S64. Compare the actual state evolution path with the operation interaction influence path to identify the deviation between the actual effect and the predicted interaction effect; S65. When there is a significant deviation between the actual state evolution path and the operation interaction path, decompose the independent contribution of each non-standard operation to the short-term disturbance and long-term positive impact on the power grid. S66. Based on the independent contributions derived from the decomposition, quantify the short-term disturbances caused by each of the non-standard operations and their long-term positive impacts to obtain the evaluation results.

5. The employee point selection intelligent analysis method of claim 4, wherein, The key operating parameters include grid voltage, frequency, and power flow distribution.

6. The employee point selection intelligent analysis method of claim 4, wherein, The specific steps in step S65 include: When there is a significant deviation between the actual state evolution path and the operation interaction path, power system analysis techniques are used to reverse analyze the physical causes of the deviation, decompose the independent contribution of each non-standard operation to the short-term disturbance and long-term positive impact on the power grid, and identify the physical enhancement, offsetting or secondary effects between the non-standard operations.

7. The intelligent analysis method for employee point evaluation according to claim 6, characterized in that, The steps of using power system analysis techniques to reverse-engineer the physical causes of the deviations and decompose the independent contribution of each of the non-standard operations to the short-term disturbances and long-term positive effects on the power grid include: S65A1. Acquire power grid topology and real-time operation data; the real-time operation data includes key power grid operation parameters; S65A2. Based on the power grid topology and the real-time operating data, and combined with the preset electrical characteristic association rules of power grid components, the physical propagation path of the deviation is obtained by tracing the deviation between the actual state evolution path and the operation interaction influence path step by step. S65A3. Identify the key physical nodes or branches that cause the deviation along the physical propagation path; S65A4. Based on the degree of impact of the non-standard operation on the critical physical node or branch, decompose the independent contribution of each non-standard operation to the short-term disturbance and long-term positive impact on the power grid.

8. The intelligent analysis method for employee point evaluation according to claim 6, characterized in that, The steps for identifying the physical enhancements, cancellations, or secondary effects between the non-standard operations include: S65B1. Collect time-series data of key power grid operating parameters before and after non-standard operations occur, as well as dispatcher operation logs; S65B2. Using event sequence analysis technology, combined with the time series data of the key operating parameters of the power grid and the dispatcher's operation log, identify the nonlinear correlation pattern between power grid state changes and multiple non-standard operations within a specific time window; S65B3. Based on the nonlinear correlation pattern, identify the physical enhancement, cancellation, or secondary effects between the nonstandard operations.

9. The intelligent analysis method for employee point evaluation according to claim 1, characterized in that, Step S7, the step of determining the operational nature of the non-standard operational behavior based on the contribution score and the evaluation result, includes: S71. Based on the relative importance of the contribution score, the short-term disturbance caused by the non-standard operating behavior, and the long-term positive impact it brings, set their respective weighting coefficients; S72. Based on the weighting coefficients, the contribution score, the short-term disturbance caused by the non-standard operating behavior, and its long-term positive impact are weighted and comprehensively calculated to obtain a comprehensive evaluation index of the non-standard operating behavior. S73. Determine the operational nature of the non-standard operating behavior based on the numerical range of the comprehensive evaluation index.

10. An intelligent analysis system for employee score selection, characterized by, include: The acquisition module is used to acquire real-time change data of key operating parameters of the power grid; The identification module is used to identify non-standard operational behaviors performed by the dispatcher; The non-standard operating behavior refers to instructions or adjustments that are not included in the preset standard operating procedures. The generation module is used to generate an expected recovery path for the power grid fault scenario when the non-standard operating behavior occurs, by referring to the power grid recovery performance under standard operating procedures in historical fault events. The monitoring module is used to track the actual changes in key power grid operating parameters after the non-standard operating behavior occurs, and to obtain the actual recovery path; The comparison module is used to compare the actual recovery path with the expected recovery path, and when it is determined that the actual recovery path is better than the expected recovery path, calculate the contribution score of the non-standard operating behavior to the improvement of power grid stability and efficiency; The evaluation module is used to evaluate the short-term disturbances caused by the non-standard operating behavior and its long-term positive impacts, and to obtain the evaluation results. The judgment module is used to determine the operational nature of the non-standard operation behavior based on the contribution score and the evaluation result, and to calculate the employee points of the corresponding dispatcher based on the determined operational nature.

Images