Power grid enterprise emergency plan system evaluation method, system, medium and equipment

Abstract

The invention discloses a power grid enterprise emergency plan system assessment method and system, a medium and equipment, and belongs to the technical field of emergency plan system assessment, and the power grid enterprise emergency plan system assessment method comprises the steps: constructing an emergency plan assessment index system; key risk indexes are identified through a fuzzy Delphi method; carrying out dynamic weight distribution on the key risk indexes based on a Bayesian best worst method; a fuzzy comprehensive evaluation method is combined with the maximum membership degree principle, and comprehensive evaluation is carried out on the emergency plan system; constructing a comprehensive risk early warning model, and calculating an early warning value of each index and an overall early warning level; and outputting an evaluation result and an early warning level. According to the method, comprehensiveness, systematicness, feasibility and practicability of emergency plan system evaluation indexes can be ensured, and emergency capability construction of power grid enterprises can be objectively and effectively evaluated.

Description

Technical Field

[0001] This invention belongs to the field of emergency response plan evaluation technology, specifically relating to a method, system, medium and equipment for evaluating the emergency response plan system of power grid enterprises. Background Technology

[0002] The statements herein provide only background information in relation to this invention and do not necessarily constitute prior art.

[0003] As a crucial link in energy consumption, urban power grids face new demands on the reliability and comprehensiveness of emergency response plan development and management by power grid companies.

[0004] Current research on the evaluation of power grid emergency response plans often employs static weight allocation (such as AHP) and deterministic evaluation methods, which struggle to effectively quantify the dynamic correlation of multi-dimensional risks and the ambiguity of expert judgments under high-proportion renewable energy integration. Traditional weighting methods (such as entropy weighting and CRITIC) fail to capture the nonlinear coupling relationships between plan elements and suffer from insufficient aggregation of subjective biases in cross-departmental expert consensus modeling. While conventional fuzzy evaluation can handle qualitative indicators, it lacks a Bayesian dynamic correction mechanism for the evolution of risk probabilities (such as the probability of cascading failures caused by wind and solar fluctuations). Although Bayesian networks and BWM methods have been applied in single domains, an evaluation framework integrating Bayesian probability updates, fuzzy linguistic variables, and multi-criteria group decision-making has not yet been formed, resulting in insufficient accuracy in evaluating the adaptability of plans for new risk scenarios such as the failure of coordinated "source-grid-load-storage" systems and the superposition of extreme weather and abnormal renewable energy output. Therefore, it is urgent to construct an evaluation paradigm that integrates dynamic weight optimization and uncertainty modeling to support the iterative upgrading of resilience-oriented emergency response plans.

[0005] Existing technologies propose a method for assessing the emergency response capabilities of power grid enterprises, including a comprehensive assessment system aimed at addressing the pain points of traditional emergency management, which relies heavily on experience and lacks quantitative tools. The hierarchical indicator system, framed by a "one plan, three systems" framework, establishes four primary indicators, thirteen secondary indicators, and thirty-eight tertiary indicators, categorized by attributes as static and dynamic, covering the entire lifecycle of emergency management. However, the inventors have found that this method has fixed scores at each stage and a rigid weighting distribution, failing to consider the risk differences among power grid enterprises of different regions or sizes. This could lead to insufficient preparation for high-risk enterprises and wasted resources for low-risk enterprises. Furthermore, its dynamic assessment requires "on-site verification by an external expert group," placing excessive reliance on experts, while in practice, expert resources are scarce, cross-regional assessments are costly, and there is significant subjective risk. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, system, medium and equipment for evaluating the emergency response plan system of power grid enterprises. This method can ensure the comprehensiveness, systematicness, feasibility and practicality of the evaluation indicators of the emergency response plan system, and can objectively and effectively evaluate the emergency response capability construction of power grid enterprises.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solution: Firstly, the technical solution of the present invention provides a method for evaluating the emergency response plan system of a power grid enterprise, including: Construct an emergency response plan evaluation indicator system; Key risk indicators were identified using the fuzzy Delphi method. Dynamic weight allocation of key risk indicators is performed based on the Bayesian best-worst method. The fuzzy comprehensive evaluation method combined with the principle of maximum membership is used to comprehensively evaluate the emergency response plan system; Construct a comprehensive risk early warning model to calculate the early warning values ​​of each indicator and the overall early warning level; Output the assessment results and warning level.

[0008] In at least one embodiment, the emergency response plan evaluation index system includes three levels of indicators; among which, the first-level indicators include the comprehensiveness of the plan analysis, the pertinence of the plan management, and the timeliness of the plan improvement. Secondary indicators include analysis of missing information, full lifecycle management, review, and updating; The three-level indicators include power grid operation risks, personal and equipment safety, high-risk and important users, natural and social factors, overall contingency plan preparation, special contingency plan setup, on-site response plans, contingency plan review and filing, and contingency plan revision and updating.

[0009] In at least one embodiment, key risk indicators are identified using the fuzzy Delphi method, specifically including: constructing a triangular fuzzy number based on expert scoring; calculating grayscale interval values ​​and consistency indicators; and identifying key risk indicators through consistency checks.

[0010] In at least one embodiment, the Bayesian best-worst method is used to dynamically assign weights to key risk indicators, specifically including: determining the risk indicators with the greatest and least impact on the emergency response plan system; constructing comparison vectors for the indicators with the greatest and least impact on the emergency response plan system; constructing multinomial probability distribution functions of the indicators based on the Bayesian best-worst method and calculating the probability of occurrence of the indicators to determine the indicator weights of the emergency response plan system.

[0011] In at least one embodiment, a fuzzy comprehensive evaluation method combined with the maximum membership principle is used to comprehensively evaluate the emergency response plan system. Specifically, this includes: constructing a factor set and a judgment set for the emergency response plan system; evaluating each factor in the factor set individually and determining its membership degree in different evaluation levels in the judgment set; and selecting the evaluation level with the highest membership degree for each evaluation object as the final evaluation result.

[0012] In at least one embodiment, the comprehensive risk warning model is specifically represented as follows:

[0013] In the formula, Indicates the first One indicator; This indicates the total number of indicators in the emergency response plan system; This indicates the warning value for a single indicator; Indicates the first The weight of each indicator; In at least one embodiment, the risk warning level includes five levels: Level I Red Warning, Level II Orange Warning, Level III Yellow Warning, Level IV Blue Warning, and Level V Green No Warning.

[0014] Secondly, the technical solution of the present invention also provides an evaluation system for the emergency response plan system of power grid enterprises, including: The system construction module is configured to: construct an emergency response plan evaluation indicator system; The key indicator identification module is configured to identify key risk indicators using the fuzzy Delphi method. The weight allocation module is configured to dynamically allocate weights to key risk indicators based on the Bayesian best-worst method. The comprehensive evaluation module is configured to use fuzzy comprehensive evaluation method combined with the maximum membership principle to conduct a comprehensive evaluation of the emergency response plan system. The early warning level calculation module is configured to: construct a comprehensive risk early warning model, calculate the early warning values ​​of each indicator and the overall early warning level; The output module is configured to output the evaluation results and warning levels.

[0015] Thirdly, the technical solution of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the power grid enterprise emergency response plan system evaluation method described in the first aspect.

[0016] Fourthly, the technical solution of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the power grid enterprise emergency plan system evaluation method described in the first aspect.

[0017] The beneficial effects of the above-described technical solution of the present invention are as follows: This invention provides an evaluation method for the emergency response plan system of power grid enterprises. It comprehensively considers multiple influencing factors and introduces the fuzzy Delphi method to identify key risk indicators of the emergency response plan, ensuring the reliability of problem identification in the emergency response plan system. In the emergency response plan evaluation system, the Bayesian best-worst-slowest method is used to design the weights of each factor, improving the rationality of the analysis. Simultaneously, fuzzy comprehensive evaluation and the principle of maximum membership are adopted to establish a comprehensive risk early warning model for the emergency response plan system, making the early warning results closer to reality. This ensures the comprehensiveness, systematicness, feasibility, and practicality of the evaluation indicators, enabling an objective and effective evaluation of the emergency response capability construction of power grid enterprises. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0019] Figure 1 This is a schematic diagram of an evaluation method for an emergency response plan system of a power grid enterprise disclosed in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram illustrating the characteristic problems of the emergency response plan preparation system for power grid enterprises disclosed in Embodiment 1 of the present invention; Figure 3 This is a flowchart of emergency plan problem identification based on the fuzzy Erfel method disclosed in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the risk warning results of the emergency response plan system in the empirical analysis disclosed in Embodiment 1 of the present invention. Detailed Implementation

[0020] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] Example 1 Driven by green and low-carbon goals, the global energy structure is undergoing profound changes, with the large-scale development and utilization of renewable energy becoming crucial for reducing greenhouse gas emissions and achieving energy structure transformation. However, with the increasing complexity of power grid structures and the growing installed capacity of renewable energy, power grid operation faces numerous practical difficulties, including poor system stability, challenges in ensuring power supply, and escalating operational risks. These challenges pose significant hurdles to the development and management of emergency response plans for power grid companies. Power grid companies play a vital role in national development, and accurately identifying problems in their emergency response plans, appropriately assessing their severity, and revising them promptly are essential means for ensuring emergency response and safety.

[0022] To promote the comprehensive improvement of the emergency response plan system of power grid enterprises, and in view of the accuracy, rationality and timeliness of power grid emergency response plans, this invention establishes a risk management assessment strategy for power grid enterprise emergency response plans based on a multi-dimensional fuzzy evaluation method, aiming to provide a certain reference for power grid enterprises to improve and perfect their existing systems.

[0023] Based on this, in a typical embodiment of the present invention, such as Figure 1 As shown in the figure, this embodiment discloses a method for evaluating the emergency response plan system of a power grid enterprise, including the following steps: S1. Construct an emergency response plan evaluation indicator system; S2. Identify key risk indicators using the fuzzy Delphi method; S3. Dynamically assign weights to key risk indicators based on the Bayesian best-worst method; S4. The emergency response plan system is comprehensively evaluated using the fuzzy comprehensive evaluation method combined with the principle of maximum membership. S5. Construct a comprehensive risk early warning model, calculate the early warning values ​​of each indicator and the overall early warning level; S6. Output the assessment results and warning level.

[0024] This method comprehensively considers multiple influencing factors. It identifies key risk indicators for emergency response plans by introducing the fuzzy Delphi method. Within the emergency response plan evaluation system, it uses the Bayesian best-worst-slowest method to design the weights of each factor. Simultaneously, it employs fuzzy comprehensive evaluation and the maximum membership principle for comprehensive assessment and construction of a comprehensive risk early warning model. This makes the early warning results closer to reality, ensuring the comprehensiveness, systematicity, feasibility, and practicality of the evaluation indicators. It can objectively and effectively evaluate the emergency response capabilities of power grid enterprises. The following section provides a detailed explanation of the above-mentioned evaluation method for the emergency response plan system of power grid enterprises, with specific implementation details.

[0025] S1. Construct an evaluation index system for emergency response plans.

[0026] This step analyzes the characteristic problems existing in the emergency response plan system, extracts key indicators representing the reliability boundary of the system from massive operational information. These indicators can profoundly reflect the inherent problems and weaknesses of the plan preparation system, provide effective input for subsequent problem level assessment, and thus guide the improvement and optimization of the plan preparation system.

[0027] The current emergency response plan development system of power grid enterprises has the following characteristic problems: Figure 2 As shown, the main problems are reflected in the following three aspects: (1) weak data foundation, and the management of emergency plans is superficial; (2) outdated methodology, and the plan evaluation system lacks scientificity; (3) disconnected from reality, and the characteristics of the problems fail to effectively guide the preparation of the plan.

[0028] To reasonably quantify and effectively characterize the operational risks arising from the non-standard preparation and management of emergency response plans by power grid enterprises, this step analyzes the characteristic problems existing in the current emergency response plan preparation system, such as formalism, lack of supervision, and poor operability, based on the multi-factor characteristics of the existing emergency response plan preparation. An emergency response plan evaluation index system is constructed, mainly reflecting the comprehensiveness, relevance, and timeliness of the emergency response plan, forming a three-level index, as shown in Table 1: Table 1 Evaluation Indicators for Emergency Response Plans of Power Grid Enterprises

[0029] S2. Identify key risk indicators using the fuzzy Delphi method.

[0030] The traditional Delphi method is complex and inefficient, making it highly unsuitable for emergency response systems requiring rapid response. Furthermore, the Delphi method heavily relies on expert subjective judgment, and factors such as expert knowledge, experience, evaluation criteria, and personal preferences can all influence the accuracy of predictions, often resulting in inconsistent outcomes. Therefore, this step employs the fuzzy Delphi method, introducing fuzzy theory into the traditional Delphi method. By using membership functions to replace expert opinions, experts are no longer required to repeatedly revise their views. The fuzzy Delphi method utilizes statistical analysis and fuzzy computation to transform expert subjective opinions into quasi-objective data, comprehensively considering the uncertainty and fuzziness of expert subjective thinking, thus improving the accuracy of factor selection and indicator selection. The specific implementation process of this step is as follows: Figure 3 As shown.

[0031] In this step, experts are first invited to evaluate each specific issue factor that needs to be assessed in the emergency response plan system of step S1, and to determine each indicator. The conservative and optimistic values ​​typically range from 0 to 10. A triangular fuzzy number is constructed based on expert scores, collecting the maximum and minimum scores for each indicator and calculating their geometric mean.

[0032] Specifically, obtaining expert opinions on the issue Minimum conservative value and maximum conservative value Then calculate the geometric mean of the conservative values. This is used as the upper limit reference value of the conservative value range.

[0033] At the same time, obtain expert opinions on the issue Minimum optimism and maximum optimism Then calculate the geometric mean of the optimistic values. This is used as the upper bound reference value for the optimistic value range.

[0034] Based on the maximum and minimum values ​​of the conservative and optimistic values, and their geometric mean, triangular fuzzy numbers are constructed for the conservative and optimistic values, respectively, and are specifically represented as follows:

[0035]

[0036] In the formula, This represents a conservative triangular fuzzy number; This represents a fuzzy number representing an optimistic value.

[0037] Then, the problem Geometric mean of conservative values With minimum optimism In comparison, if This indicates that the experts' scores for this indicator are consistent. The consistency of the indicator is calculated as follows:

[0038] like If the score is 0, it indicates that experts disagree on the score of this indicator, and it is necessary to calculate the grayscale interval value to reflect the degree of disagreement among experts on this indicator.

[0039] The specific calculation method for the grayscale interval value is expressed as follows:

[0040] In the formula, Reflecting on the experts' views on the issue The degree of disagreement between conservative and optimistic values; This reflects the degree of disagreement among experts regarding the optimism level.

[0041] like This indicates that the experts' scores for this indicator are consistent. The consistency of the indicator is calculated as follows:

[0042] like If the score is negative, it indicates that the experts' evaluations of the indicator are inconsistent, and the experts need to re-score the indicator.

[0043] When experts agree on the score for an indicator, a consistency test is used to identify key risk indicators. Specifically, the consistency value of each indicator is compared with a set threshold. Indicators with a consistency value greater than the set threshold are retained as key risk indicators, while indicators with a consistency value less than the set threshold are removed. In this embodiment, the set threshold is 5.5.

[0044] S3. Dynamically assign weights to key risk indicators based on the Bayesian best-worst method.

[0045] Traditional analytic hierarchy process (AHP) relies on subjective judgment, which becomes complex when there are too many indicators involved, making the construction of the judgment matrix and consistency checks complicated. The best-worst method (BWM) can help reduce the number of comparisons between indicators, but it suffers from anomaly sensitivity and limited information provision. In 2019, Mohammadi and Rezaei combined Bayesian theory with the traditional BWM method to form the Bayesian BWM method, which can solve the above problems. Furthermore, the Bayesian method supports dynamic updates of posterior probabilities, allowing weights to continuously reflect the latest information, making it suitable for emergency plans that need to be adjusted according to environmental changes. Therefore, this step uses Bayesian BWM to design weights for problem indicators in the power grid enterprise's emergency plan system and dynamically allocates weights to key risk indicators. The basic steps are as follows: First, based on the emergency response plan problem identification results from the fuzzy Delphi method, the consistency values ​​calculated for each indicator are compared using the fuzzy Delphi method. The magnitude of the impact is determined by the numerical value, where the consistency value... The highest value is determined to be the risk indicator that has the greatest impact on the emergency response plan system. Consistency value The lowest value is determined as the risk indicator with the least impact on the emergency response plan system. .

[0046] Then, a comparison vector is constructed, which includes the risk indicators with the greatest impact. Other indicators For comparison, based on the Saaty scaling method, 1-9 is used as the numerator. Other indicators Importance, 1 represents and Equal importance, 9 represents Compare Much more important. Construct risk indicators with minimal impact using the same approach. The comparison vectors yield the comparison vectors of the factors with the greatest influence. Comparison vector with the least influential factor Specifically, it is expressed as:

[0047] Then, based on the Bayesian best-worst method, a multinomial probability distribution function for the index is constructed, with the worst-case index (i.e., the risk index pre-selected in the Bayesian best-worst method that has the least impact on the emergency response plan system) being selected. For example, construct The polynomial probability distribution function is expressed as:

[0048] In the formula, Indicates the first The weights of each indicator, and satisfying .

[0049] Finally, the probability of the indicators occurring is calculated to determine the indicator weights in the emergency response plan system. This step transforms the traditional weight determination process into a probability distribution estimation problem, using a hierarchical Bayesian model to estimate the comparison vectors of multiple experts (e.g., ...). and Using observational data, we can jointly estimate an indicator weight vector that reflects group consensus. Meanwhile, the uncertainty of the estimate is quantified, and the consistency of the index is checked by calculating the degree of matching between the expert comparison vector and the optimal weight vector under the posterior probability based on the Bayesian posterior distribution.

[0050] S4. The emergency response plan system is comprehensively evaluated by using the fuzzy comprehensive evaluation method combined with the principle of maximum membership.

[0051] After calculating the weights of the factors involved in the risk using the Bayesian best-worst method, this step employs a fuzzy comprehensive evaluation model and the maximum membership principle to assess the emergency response plan system. The fuzzy evaluation method is a multi-objective decision-making method based on fuzzy mathematics theory for prediction and evaluation. Its algorithm is similar to human thinking patterns, describing objects according to degree, and is suitable for handling multi-factor, multi-level evaluation problems of complex systems. Based on this, this step uses the fuzzy comprehensive evaluation method to achieve a three-dimensional diagnosis of the emergency response plan system. Its basic steps are as follows: First, identify all risk indicators affecting the evaluation object, and then group these risk indicators into a factor set, specifically represented as follows:

[0052] Then, a set containing all possible evaluation results is constructed, denoted as the evaluation set. In this embodiment, the evaluation results of the emergency response plan evaluation system include four categories: major risk, significant risk, general risk, and minor risk. Therefore, the evaluation set in this step can be specifically represented as follows:

[0053] In the formula, This indicates that there is a significant risk; This indicates that there is a significant risk. This indicates the presence of general risk; This indicates that there is a relatively small risk.

[0054] Next, each factor in the factor set is evaluated individually to determine its membership degree at different evaluation levels. Membership degree Indicates the first The factor in the first The degree of compliance in each evaluation level is typically represented by a value between 0 and 1.

[0055] Following the principle of maximum membership, the evaluation level with the highest membership degree is selected as the final evaluation result for each evaluated object, transforming qualitative judgments into a basis for scientific decision-making. Evaluation Matrix Specifically, it is expressed as follows:

[0056] Combined with evaluation matrix and weight vector The comprehensive evaluation result vector is calculated and obtained, specifically represented as follows:

[0057] In the formula, The comprehensive evaluation result vector is determined by finding the element with the largest value and mapping it to the preset evaluation set level based on its position in the vector (e.g., the first element represents "major risk"). In other words, after calculating the comprehensive evaluation result vector B, the position corresponding to its maximum value determines the risk level (major, relatively large, general, relatively small).

[0058] S5. Construct a comprehensive risk early warning model and calculate the early warning values ​​of each indicator and the overall early warning level.

[0059] In traditional emergency response plan risk assessments, even if the risk value of one indicator reaches its maximum, the overall risk may still be reduced because the risks of other indicators are relatively low. However, in the actual construction and operation of emergency response plan systems in power grid companies, if an indicator poses an extremely serious risk, the risk it brings cannot and should not be ignored. Therefore, in constructing the comprehensive risk early warning model for emergency response plans in this step, we assume that the indicator... The risk is , The risk warning threshold is Then, individual indicators. warning value It can be represented as:

[0060] If the indicator Risk Not greater than the threshold If the early risk warning value is 0, it indicates that the indicator... There was no early risk warning. Conversely, if the indicators... Risk Greater than the threshold If the early warning value is greater than 0, then the early risk warning value will be greater than 0. Based on the warning value of a single indicator, the comprehensive risk warning model for engineering projects can be expressed as:

[0061] In the formula, The base is ; This represents the total number of key risk indicators involved in the comprehensive risk warning calculation. Based on equation (13), the comprehensive risk warning value of this emergency plan system can be calculated.

[0062] S6. Output the assessment results and warning level.

[0063] In this step, the risk level warning includes five levels: Level I Red Warning, Level II Orange Warning, Level III Yellow Warning, Level IV Blue Warning, and Level V Green No Warning. The warning criteria are shown in Table 2.

[0064] Table 2 Risk Level Warning Standards

[0065] The indicator value is a percentage score based on the linear transformation of the warning value. The warning value of the individual indicator of the selected emergency plan is calculated according to formula (12), and the comprehensive risk warning value of the selected emergency plan is calculated according to formula (13). The warning level of the individual indicator and the comprehensive risk warning level of the selected emergency plan can be obtained from Table 2.

[0066] To verify the effectiveness of the power grid enterprise emergency response plan system evaluation strategy in this method, this embodiment conducted an empirical analysis. The project information came from a prefecture-level city power grid, and a representative emergency response plan system was selected from the project database of that region as a case for evaluation. Based on the results, corresponding suggestions were proposed.

[0067] First, after identifying the risk factors of a power grid with a high proportion of renewable energy, the key risk factors are determined according to the fuzzy Delphi method, and the risk identification results are obtained through steps S1 and S2, as shown in Table 3.

[0068] Table 3. Results of Emergency Response Plan Problem Identification Based on Fuzzy Delphi Method

[0069] It can be seen that C7 has the highest consistency value, reflecting the widespread "theoretical" nature of current contingency plans, with a lack of operational guidance for specific scenarios in on-site response measures. C5, on the other hand, has a lower consistency value, indicating fundamental disagreements among experts regarding the overall quality and structural rationality of the contingency plan. This reflects significant differences in the planning philosophies and methodologies among different units. Based on the consistency values ​​in Table 3, C1 to C9 are categorized into three types of problems, as shown in Table 4.

[0070] Table 4 Consistency Analysis

[0071] Through the consistency analysis of Tables 3 and 4, the data can be used to quantify the severity of the problem. Power grid companies can realize the transformation of emergency plan evaluation from "experience-driven" to "data-driven". This can help decision-makers understand the impact of different problem factors on the emergency plan system, significantly improve the practical effectiveness of the plan system, and formulate more accurate emergency plan management strategies and preventive measures.

[0072] Based on the above indicators, the key risk indicators were dynamically weighted according to the Bayesian best-worst method, and the results are shown in Table 5.

[0073] Table 5. Risk Indicator Weights in the Emergency Response Plan System

[0074] Analysis of Table 5 shows that the top three weighted indicators are C1, C7, and C9, which are the same as the key risk factors identified by the fuzzy Delphi method. Among them, the weight of power grid operation risk (C1) is 0.1976, indicating that power grid operation risk is considered the most critical consideration in the emergency response plan system, directly related to the reliability and security of power supply. The weights of on-site handling plans (C7) and plan revision and updates (C9) follow closely behind, at 0.1812 and 0.1782 respectively. This reflects that specific and operable on-site handling plans in emergency response plans can quickly control the development of the situation, and the dynamic management and adaptive adjustment of the plans are also highly valued, requiring timely updates to maintain their effectiveness and applicability.

[0075] Because the safety of power grid enterprises is crucial to the economy and people's lives, the probability of activating various emergency plans is relatively low, resulting in a very small sample size in historical data, making it impossible to simply use accuracy to measure the training effect. According to Table 1, the samples are divided into three major categories and nine subcategories. However, for the power system, the cost of misclassifying these samples by the model varies. Errors can significantly impact the emergency safety of power grid enterprises. To address this, a cost-sensitive learning method is employed to reduce the impact of imbalanced samples. Based on the problem identification results and the weights of each factor, fuzzy comprehensive evaluation is used to assess the emergency plan system cases. The evaluation matrix results are shown in Table 6.

[0076] Table 6 Risk Indicator Assessment Matrix for Emergency Response Plan System

[0077] Calculations show that the overall risk membership of this emergency response plan system is: Based on the principle of maximum membership, the risk level is 0.411, indicating that the emergency response plan system has a significant risk and a high probability of high-concern issues occurring. Therefore, power grid companies need to prioritize taking corresponding measures for high-concern issues C1, C7, and C9 and promptly revise and improve the current emergency response plan system.

[0078] Finally, the warning values ​​and levels of individual indicators for the selected emergency response plan, as well as the overall risk warning results, are calculated. Figure 4 As shown in the table, according to the risk level early warning standard table, the overall risk early warning level of the power grid company's emergency plan system is Level II (orange), indicating that the system has a high risk and a special emergency response mechanism needs to be activated immediately. Among them, indicators C1, C7, and C9 are Level III (yellow), and C8 is Level IV (blue). Although no individual indicator has reached Level I or Level II warning, indicating that the current risk is still under control, these Level III and IV warning indicators reveal weak links within the system. Under specific disturbances, there is a possibility of risk escalation and transmission, but the local risks existing in this plan cannot be ignored.

[0079] Example 2 In a typical embodiment of the present invention, this embodiment discloses an emergency response plan system evaluation system for power grid enterprises, comprising: The system construction module is configured to: construct an emergency response plan evaluation indicator system; The key indicator identification module is configured to identify key risk indicators using the fuzzy Delphi method. The weight allocation module is configured to dynamically allocate weights to key risk indicators based on the Bayesian best-worst method. The comprehensive evaluation module is configured to use fuzzy comprehensive evaluation method combined with the maximum membership principle to conduct a comprehensive evaluation of the emergency response plan system. The early warning level calculation module is configured to: construct a comprehensive risk early warning model, calculate the early warning values ​​of each indicator and the overall early warning level; The output module is configured to output the evaluation results and warning levels.

[0080] Example 3 In a typical embodiment of the present invention, this embodiment provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the steps in the power grid enterprise emergency response plan system evaluation method described in Embodiment 1. These steps include: S1. Construct an emergency response plan evaluation indicator system; S2. Identify key risk indicators using the fuzzy Delphi method; S3. Dynamically assign weights to key risk indicators based on the Bayesian best-worst method; S4. The emergency response plan system is comprehensively evaluated using the fuzzy comprehensive evaluation method combined with the principle of maximum membership. S5. Construct a comprehensive risk early warning model, calculate the early warning values ​​of each indicator and the overall early warning level; S6. Output the assessment results and warning level.

[0081] Example 4 In a typical embodiment of the present invention, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the power grid enterprise emergency response plan system evaluation method described in Embodiment 1. These steps include: S1. Construct an emergency response plan evaluation indicator system; S2. Identify key risk indicators using the fuzzy Delphi method; S3. Dynamically assign weights to key risk indicators based on the Bayesian best-worst method; S4. The emergency response plan system is comprehensively evaluated using the fuzzy comprehensive evaluation method combined with the principle of maximum membership. S5. Construct a comprehensive risk early warning model, calculate the early warning values ​​of each indicator and the overall early warning level; S6. Output the assessment results and warning level.

[0082] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. 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. A method for evaluating the emergency response plan system of power grid enterprises, characterized in that, include: Construct an emergency response plan evaluation indicator system; Key risk indicators were identified using the fuzzy Delphi method. Dynamic weight allocation of key risk indicators is performed based on the Bayesian best-worst method. The fuzzy comprehensive evaluation method combined with the principle of maximum membership is used to comprehensively evaluate the emergency response plan system; Construct a comprehensive risk early warning model to calculate the early warning values ​​of each indicator and the overall early warning level; Output the assessment results and warning level.

2. The method for evaluating the emergency response plan system of a power grid enterprise as described in claim 1, characterized in that, The emergency response plan evaluation indicator system includes three levels of indicators; among them, the first level indicators include the comprehensiveness of the plan analysis, the pertinence of the plan management, and the timeliness of the plan improvement. Secondary indicators include analysis of missing information, full lifecycle management, review, and updating; The three-level indicators include power grid operation risks, personal and equipment safety, high-risk and important users, natural and social factors, overall contingency plan preparation, special contingency plan setup, on-site response plans, contingency plan review and filing, and contingency plan revision and updating.

3. The method for evaluating the emergency response plan system of a power grid enterprise as described in claim 1, characterized in that, The key risk indicators are identified using the fuzzy Delphi method, which includes: constructing a triangular fuzzy number based on expert scoring; calculating gray-scale interval values ​​and consistency indicators; and identifying key risk indicators through consistency testing.

4. The method for evaluating the emergency response plan system of a power grid enterprise as described in claim 1, characterized in that, The Bayesian best-worst method is used to dynamically assign weights to key risk indicators. Specifically, this includes: identifying the risk indicators with the greatest and least impact on the emergency response plan system; constructing comparison vectors for the indicators with the greatest and least impact on the emergency response plan system; constructing multinomial probability distribution functions for the indicators based on the Bayesian best-worst method and calculating the probability of occurrence of the indicators to determine the indicator weights of the emergency response plan system.

5. The method for evaluating the emergency response plan system of a power grid enterprise as described in claim 1, characterized in that, The fuzzy comprehensive evaluation method, combined with the principle of maximum membership, is used to comprehensively evaluate the emergency response plan system. Specifically, this includes: constructing a factor set and a judgment set for the emergency response plan system; evaluating each factor in the factor set individually and determining its membership degree in different evaluation levels in the judgment set; and selecting the evaluation level with the highest membership degree for each evaluation object as the final evaluation result.

6. The method for evaluating the emergency response plan system of a power grid enterprise as described in claim 1, characterized in that, The comprehensive risk early warning model is specifically represented as follows: In the formula, Indicates the first One indicator; This indicates the total number of indicators in the emergency response plan system; This indicates the warning value for a single indicator; Indicates the first The weight of each indicator.

7. The method for evaluating the emergency response plan system of a power grid enterprise as described in claim 1, characterized in that, The risk warning levels are divided into five levels: Level I (Red), Level II (Orange), Level III (Yellow), Level IV (Blue), and Level V (Green, No Warning).

8. An evaluation system for emergency response plans of power grid enterprises, characterized in that, include: The system construction module is configured to: construct an emergency response plan evaluation indicator system; The key indicator identification module is configured to identify key risk indicators using the fuzzy Delphi method. The weight allocation module is configured to dynamically allocate weights to key risk indicators based on the Bayesian best-worst method. The comprehensive evaluation module is configured to use fuzzy comprehensive evaluation method combined with the maximum membership principle to conduct a comprehensive evaluation of the emergency response plan system. The early warning level calculation module is configured to: construct a comprehensive risk early warning model, calculate the early warning values ​​of each indicator and the overall early warning level; The output module is configured to output the evaluation results and warning levels.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the power grid enterprise emergency response plan system evaluation method as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the power grid enterprise emergency response plan system evaluation method as described in any one of claims 1-7.

Images