Multi-dimensional electricity market operation effect evaluation method and system

Abstract

The invention discloses a multi-dimensional electric power market operation effect evaluation method and system, and relates to the technical field of electric power market operation effect evaluation, and the method comprises the steps: constructing a multi-dimensional index system covering the core operation dimension of an electric power market, and according to a combined weighting model composed of a fusion analytic hierarchy process and an entropy weight method, carrying out the evaluation of the operation effect of the electric power market; subjective experience and objective data are combined; and calculating the distance between each province and an ideal solution according to the combined data by using a TOPSIS model to realize comprehensive sorting, and processing uncertainty through fuzzy logic and Bayesian reasoning to complete evaluation of the multi-dimensional power market operation effect. According to the method, subjective experience and objective data are fused through combined empowerment, inter-provincial multi-index comprehensive sorting is realized by means of TOPSIS, risk causal reasoning is performed by using a fuzzy Bayesian network (FBN), a complementary closed loop of'macroscopic sorting-microscopic diagnosis' is formed, and the blank in the field of provincial power market dynamic evaluation and risk traceability in the prior art is filled.

Description

Technical Field

[0001] This invention relates to the field of power market operation performance evaluation technology, and more specifically, to a multi-dimensional power market operation performance evaluation method and system. Background Technology

[0002] Currently, inter-provincial and intra-provincial markets operate in a coordinated manner, forming a multi-product market structure covering medium- and long-term, spot, and ancillary service transactions. With the advancement of electricity market construction in various provinces, the number of market participants, the scale of transaction volume, and the level of platform automation have significantly increased, accumulating massive amounts of transaction information and grid operation data. However, the current market at all levels lacks a unified and coordinated indicator system for data statistics and reporting, resulting in problems such as misunderstandings of business operations, inconsistent statistical standards, and incomplete indicators. A scientific evaluation system is urgently needed to support market performance analysis.

[0003] Existing power market evaluation technologies suffer from core deficiencies, including a lack of integration of financial and physical indicators in the design of the indicator system, limitations in the simplification of evaluation methods, and insufficient data processing capabilities. Current technologies have not yet formed a complete technical chain encompassing "indicator system - weighting mechanism - evaluation model - risk diagnosis," thus failing to meet the needs of dynamic evaluation and risk tracing in provincial power markets. Summary of the Invention

[0004] To address the aforementioned issues, the present invention aims to provide a multi-dimensional power market operation performance evaluation technology, which fills the gap in existing technologies for dynamic evaluation and risk tracing of provincial power markets.

[0005] To achieve the above technical objectives, this application provides a multi-dimensional method for evaluating the operational effectiveness of the electricity market, comprising the following steps: A multi-dimensional indicator system covering the core operational dimensions of the power market is constructed, and subjective experience is combined with objective data based on a combined weighting model composed of the fusion analytic hierarchy process and the entropy weight method. Using the TOPSIS model, the distance between each province and the ideal solution is calculated based on the combined data to achieve a comprehensive ranking. After handling the uncertainty through fuzzy logic and Bayesian inference, the evaluation of the multi-dimensional power market operation effectiveness is completed.

[0006] Preferably, when constructing a multi-dimensional indicator system, the system is constructed based on indicators of market structure, market behavior, and market performance.

[0007] Preferably, when obtaining the market structure indicators, the market structure indicators include: Market-based electricity trading volume ratio = Medium- and long-term electricity trading volume / Total social electricity consumption × 100%; Market-based electricity price ratio = Enterprise's transaction price per kilowatt-hour / Average market price per kilowatt-hour; Medium- to long-term (annual) supply-demand ratio = sum of the maximum declared quantities of participating market entities in the annual transaction / total annual demand in the transaction market; Spot (monthly) supply-demand ratio = (Spot supply and price quotes for each period / Total maximum declared output of generating units / Net load demand); Market power is divided into wholesale market power and retail market power. It monitors the market share and structural composition of power generation companies in the market and is calculated by weighting three market power indicators: Top-m, HHI, and RSI.

[0008] Preferably, when obtaining the market behavior component indicators, the market behavior component indicators include: Market participant participation = w1 × power generation enterprise participation + w2 × electricity sales company participation + w3 × electricity user participation, where w1 + w2 + w3 = 1 represents the different weights of the three types of market participants; Medium- and long-term transaction turnover rate = (medium- and long-term hedging transaction volume + medium- and long-term transfer transaction volume + medium- and long-term repurchase transaction volume) / medium- and long-term cumulative transaction volume × 100%; Price volatility refers to the level of fluctuation of the average market price over the most recent statistical periods. It can be calculated separately for wholesale and retail markets using the coefficient of variation. Relative Strength Index of Electricity Prices ; Peak-valley electricity price response coefficient In the formula, RC is the response coefficient, △Ep is the peak power generation increment, △Ev is the valley power generation decrease, and Sp is the peak-valley price difference. Demand response volatility In the formula, For demand response volatility, The standard deviation of the response quantity This represents the average response value.

[0009] Preferably, when obtaining the market performance indicators, the market performance indicators include: Medium- and long-term contract signing rate = (Medium- and long-term transaction volume for different target periods / Total medium- and long-term transaction volume) × 100%; Impact of medium- and long-term transactions = Medium- and long-term contract electricity volume × Medium- and long-term contract electricity price / Settlement electricity cost × 100%; Green certificate turnover rate In the formula, TUR is the green certificate turnover rate, Qt is the green certificate trading volume, and Qh is the green certificate holding volume; New energy utilization rate = (Total new energy power generation - Curtailed wind and solar power) / Total new energy power generation × 100%; Green electricity premium In the formula, PR is the green electricity premium rate, Pg is the green electricity transaction price, and Pc is the coal-fired power benchmark price. The peak-valley difference rate for spot market transactions = (maximum spot clearing electricity price - minimum spot clearing electricity price) / (upper limit spot clearing price - lower limit spot clearing price) × 100%, calculated based on the average daily peak-valley price difference during the statistical period; Load peak-valley difference rate = (net maximum load - net minimum load) / net maximum load × 100%, calculated based on the daily average load peak-valley difference during the statistical period, where net load = local load + externally transmitted power - externally purchased power; In the formula, Qh represents the electricity held by market participants; Qt represents the total available electricity in the market.

[0010] Preferably, when combining subjective experience with objective data, the indicators are standardized and divided into three categories: positive, negative, and interval. After eliminating the influence of dimensions using linear transformation, a judgment matrix is ​​constructed through expert questionnaires based on the analytic hierarchy process (AHP), transforming subjective experience into quantitative weights. This approach is suitable for evaluation scenarios where there is hierarchical logic among indicators, reflecting the impact of policy guidance on the importance of indicators. Furthermore, based on the entropy weight method, objective weights are assigned according to the variability of indicators, where indicators with greater data fluctuations carry more information and have higher weights.

[0011] Preferably, when performing comprehensive ranking, a combined weight is formed based on the quantitative weight transformed from subjective experience and the weight result of objective weighting of indicator variability based on the entropy weight method. By constructing a decision matrix and introducing the combined weight, an ideal solution is obtained. Then, the provinces are comprehensively ranked according to the Euclidean distance between each province and the ideal solution and the closeness obtained from the Euclidean distance.

[0012] Preferably, after comprehensive ranking, the precise indicator values ​​are mapped to fuzzy sets based on the fuzzy Bayesian Network (FBN) to construct the network topology. The causal relationship between indicators is defined through expert knowledge or data mining. The conditional probability table is trained based on historical data. After inputting the fuzzy membership degree of the indicators, the posterior probability distribution of the target node is derived through the Bayesian formula, thereby achieving in-depth tracing of market risks and identifying key risk paths.

[0013] This invention discloses a multi-dimensional electricity market operation performance evaluation system, used to implement the aforementioned multi-dimensional electricity market operation performance evaluation method, comprising: The data acquisition and processing module is used to construct a multi-dimensional indicator system covering the core operational dimensions of the power market. It combines subjective experience with objective data based on a combined weighting model composed of the analytic hierarchy process and the entropy weight method. The evaluation module uses the TOPSIS model to calculate the distance between each province and the ideal solution based on the combined data to achieve a comprehensive ranking. After handling uncertainty through fuzzy logic and Bayesian inference, it completes the evaluation of the multi-dimensional effectiveness of the power market operation.

[0014] The present invention discloses the following technical effects: This invention combines subjective experience and objective data through weighted combination, uses TOPSIS to achieve comprehensive ranking of multiple indicators among provinces, and utilizes fuzzy Bayesian networks (FBN) for risk causal reasoning to form a complementary closed loop of "macro ranking - micro diagnosis", filling the gap in existing technologies in the field of dynamic evaluation and risk tracing of provincial power markets. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is the TOPSIS sorting flowchart described in this invention; Figure 2 This is a diagram of the fuzzy Bayesian network topology described in this invention; Figure 3 This is the risk tracing result diagram of Province B as described in this invention; Figure 4 This is a schematic diagram of the method described in this invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application 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 this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0018] like Figures 1-4 As shown, this invention provides a multi-dimensional power market operation performance evaluation technology. This technology utilizes a combined model to evaluate the provincial power market from multiple dimensions, specifically including the following steps: (1) First, a multi-dimensional indicator system covering the core operational dimensions of the electricity market is constructed. The market-based price-per-kilowatt-hour ratio indicator reflects the pricing power of power generation enterprises in market competition by comparing the enterprise's transaction price per kilowatt-hour with the market average price per kilowatt-hour. When the ratio is greater than 1, it indicates that the enterprise has the ability to set a premium. The green electricity premium rate indicator quantifies the marketization degree of the environmental value of green electricity by comparing the difference between the green electricity transaction price and the coal-fired power benchmark price. When the premium rate exceeds 5%, the internal rate of return of renewable energy projects can be increased by 2-3 percentage points. The electricity retention ratio indicator measures the degree of market power concentration by comparing the electricity retained by market entities with the total available electricity. When the ratio exceeds 20%, there may be collusion or monopolistic behavior among market entities. The peak-valley electricity price response coefficient indicator assesses the unit's ability to regulate time-of-use electricity prices by dividing the difference between the increase in power generation during peak hours and the decrease in power generation during valley hours by the peak-valley price difference. When the response coefficient is greater than 0.5, the unit's peak-shaving revenue can cover the cost. The demand response volatility index quantifies the stability of demand response resources by measuring the ratio of the standard deviation of the response volume to the average response volume. When the volatility is below 15%, demand response resources can serve as a reliable peak-shaving measure. The green certificate turnover rate index measures the trading activity of the green certificate market by measuring the ratio of green certificate trading volume to holding volume. When the turnover rate exceeds 200%, it indicates the presence of speculative behavior in the green certificate market. The electricity price relative strength index measures the relative strength trend of electricity prices over a certain period by comparing the magnitude of regional electricity price increases and decreases, providing market participants with price trend references. A system based on the SCP model is constructed, consisting of 9 primary indicators and 20 secondary indicators across three dimensions: market structure, market behavior, and market performance.

[0019] (2) Secondly, the combined weighting model achieves an organic combination of subjective experience and objective data by integrating the analytic hierarchy process (AHP) and the entropy weight method. First, the indicators are standardized and divided into three categories: positive, negative, and interval. Linear transformation is used to eliminate the influence of dimensions. The AHP constructs a judgment matrix through expert questionnaires, transforming subjective experience into quantitative weights. It is suitable for evaluation scenarios where there is hierarchical logic between indicators, reflecting the influence of policy guidance on the importance of indicators. The entropy weight method objectively assigns weights based on the variability of indicators. The greater the data fluctuation, the more information the indicator carries and the higher its weight. It is suitable for multi-dimensional indicator fusion scenarios, reflecting the real impact through data fluctuations. The two are combined in a ratio of 0.4:0.6. This ratio is verified by Monte Carlo simulation of market data from five provinces from 2019 to 2024. The results show that the evaluation results under this ratio have the highest degree of consistency with the actual market operation results. It ensures the weight ratio of policy indicators and strengthens the objective influence of data-sensitive indicators, achieving a balance between policy guidance and data patterns.

[0020] (3) Furthermore, the TOPSIS inter-provincial ranking model achieves comprehensive ranking by calculating the distance between each province and the ideal solution. After constructing the decision matrix, the indicators are standardized, combined weights are introduced to construct a weighted matrix, positive and negative ideal solutions are determined, and multi-indicator comprehensive ranking of inter-provincial market performance is achieved by calculating Euclidean distance and proximity, effectively solving the problem of comprehensive comparison of multiple indicators among provinces.

[0021] (4) Finally, the fuzzy Bayesian network risk tracing model handles uncertainty through fuzzy logic and Bayesian inference. It maps precise indicator values ​​to fuzzy sets, constructs a network topology, defines causal relationships between indicators through expert knowledge or data mining, trains conditional probability tables based on historical data, and derives the posterior probability distribution of target nodes through Bayesian formulas after inputting the fuzzy membership degrees of the indicators, thereby achieving in-depth tracing of market risks and identifying key risk paths.

[0022] This invention constructs a multi-dimensional evaluation model based on the multidimensional complexity of the electricity market, the heterogeneity of indicators, and the duality of evaluation objectives. It employs a three-tiered progressive logic: "subjective-objective weighting integration → multi-attribute decision ranking → uncertainty risk diagnosis," achieving a complete analytical chain from determining indicator weights to uncovering deep-seated market issues. This model constructs a multi-dimensional indicator system encompassing market structure, market behavior, and market performance, combining a weighting mechanism of the Analytic Hierarchy Process (AHP) and entropy weighting, as well as complementary models of TOPSIS and FBN, forming an evaluation system that is both policy-guided and data-driven.

[0023] Example: The multi-dimensional power market operation effectiveness evaluation technology disclosed in this invention specifically includes the following execution process: (1) Implementation of a multi-dimensional indicator system: (a) Data acquisition and preprocessing: As shown in Figure 1: This invention collects power market operation data for provinces A, B, and C in 2024 (data source: public reports from the power trading centers of the three provinces and grid dispatch data). The reason for choosing these three provinces is that they respectively represent regions with a high degree of marketization (province A), a region with a medium degree of marketization (province B), and a region with a high proportion of renewable energy (province C), and are therefore typical and representative. Data preprocessing adopts the Z-Score standardization method.

[0024] Table 1 (b) Calculation method of core indicators: Market structure section: Market-based electricity trading volume ratio = Medium- and long-term electricity trading volume / Total social electricity consumption × 100% Market-based electricity price ratio = Enterprise's transaction price per kilowatt-hour / Average market price per kilowatt-hour Medium- to long-term (annual) supply-demand ratio = Sum of maximum declared quantities by participating market entities in the annual transaction / Total annual demand in the transaction market Spot (monthly) supply-demand ratio = Spot supply and price quotes for each period / Total maximum declared output of generating units / Net load demand Market power is divided into wholesale market power and retail market power. It mainly monitors the market share and structural composition of power generation companies in the market and is calculated by weighting three market power indicators: Top-m, HHI, and RSI.

[0025] Market behavior section: Market participant participation rate = w1 × power generation enterprise participation rate + w2 × electricity sales company participation rate + w3 × electricity user participation rate. Where w1 + w2 + w3 = 1 represents the different weights of the three types of market participants, tentatively set at 35%, 35%, and 30%.

[0026] Medium- and long-term transaction turnover rate = (Medium- and long-term hedging transaction volume + Medium- and long-term transfer transaction volume + Medium- and long-term repurchase transaction volume) / Medium- and long-term cumulative transaction volume × 100% Price volatility refers to the fluctuation level of the average market price over several recent statistical periods (e.g., the past year). It can be calculated separately for wholesale and retail markets using the coefficient of variation. Coefficient of variation .

[0027] Relative Strength Index of Electricity Prices ; Peak-valley electricity price response coefficient In the formula, RC is the response coefficient, ΔEp is the peak power generation increment, ΔEv is the valley power generation decrease, and S p This refers to the price difference between peak and off-peak periods.

[0028] Demand response volatility In the formula, For demand response volatility, The standard deviation of the response quantity This represents the average response value.

[0029] Market performance section: Medium- and long-term contract signing rate = Medium- and long-term transaction volume for different target periods (annual (multi-year), monthly (multi-month), intra-month, etc.) / Total medium- and long-term transaction volume × 100%.

[0030] Impact of medium- and long-term transactions = Medium- and long-term contract electricity volume × Medium- and long-term contract electricity price / Settlement electricity cost × 100%.

[0031] Green certificate turnover rate In the formula, TUR is the green certificate turnover rate, and Q is the green certificate turnover rate. t For green certificate trading volume, Q h This refers to the number of green certificates held.

[0032] New energy utilization rate = (Total new energy power generation - Curtailed wind and solar power) / Total new energy power generation × 100% Green electricity premium In the formula, PR is the green electricity premium rate, and P g The transaction price for green electricity, P c This is the benchmark price for coal-fired power.

[0033] The peak-valley difference rate for spot market is calculated as follows: (maximum spot clearing price - minimum spot clearing price) / (upper limit spot clearing price - lower limit spot clearing price) × 100%, based on the average daily peak-valley price difference during the statistical period.

[0034] The load peak-valley difference rate = (maximum net load - minimum net load) / maximum net load × 100%, calculated based on the average daily load peak-valley difference during the statistical period. Wherein, net load = local load + externally transmitted power - externally purchased power. In the formula, Q h Electricity reserved for market participants (available electricity not involved in trading); Q t This refers to the total available electricity in the market (the amount of electricity that can be provided by the total installed capacity of power generation companies).

[0035] (2) Implementation of the combined weighting model: (a) Standardization of indicators: The indicators are divided into three categories: positive, negative, and range. Positive indicators, such as the proportion of electricity traded through market mechanisms, have the following standardized formula: ; Contrarian indicators, such as demand response volatility, are adopted using: ; Interval indicators, such as the supply-demand ratio, are mapped to the [0,1] interval through a piecewise function to ensure that all indicators have the same dimensions.

[0036] (b) Implementation of AHP subjective empowerment: A three-tiered hierarchical structure of "market structure - price response - transaction performance" was constructed, and five experts (including representatives from regulators, power generation, and research institutions) were invited to build a judgment matrix using a 1-9 scale. Taking the market structure dimension as an example, with the ratio of "market-traded electricity volume to supply-demand ratio = 3:1", the maximum eigenvalue and the consistency index (CI) were calculated to ensure... .

[0037] The weight vector is then calculated using the square root method. For example, the AHP weight of the market-based electricity trading volume is 0.15, reflecting the policy orientation towards core indicators.

[0038] (c) Implementation of objective weighting using the entropy weight method: Construct a probability matrix based on the normalized matrix Z: ; Calculate the entropy value: ; Weight: ; Indicators with large data fluctuations, such as electricity price volatility, have low entropy values ​​( It received a high weight of 0.08.

[0039] (d) Combination weight fusion: The weights of AHP and entropy weighting are combined at a ratio of 0.4:0.6, as shown in the formula. ; For example, when the AHP weight of 0.02 for the green electricity premium rate is combined with the entropy weight weight of 0.04, the combined weight is 0.032. This weight reflects both the policy support for green energy (AHP weight) and the impact of the fluctuation of the green electricity premium rate in market data (entropy weight weight), thus achieving a balance in two dimensions.

[0040] (3) Implementation of the TOPSIS inter-provincial ranking model: (a) Decision matrix processing: An original matrix X is constructed using n provinces and m=18 indicators, which is then standardized to obtain matrix Z. Taking province A with 85% of its electricity traded through market mechanisms and province B with 60% as an example, the standardized positive indicators are 1.0 for province A, 0 for province B, and 0.48 for province C. The standardization process eliminates the influence of different indicator dimensions, ensuring the scientific nature of the ranking.

[0041] (b) Construction of the weighted matrix and ideal solution: Introducing combined weights Calculate the weighting matrix: ; Positive Ideal Solution Take the maximum value of the positive indicator (e.g., the weighted average of the market-based electricity trading volume in Province A, which is 0.138), and the negative ideal solution. Taking the minimum value (e.g., 0 for province B), the determination of positive and negative ideal solutions provides a quantitative benchmark for comparing the market performance of each province.

[0042] (c) Proximity calculation and sorting Calculate the Euclidean distance: ; ; Proximity: ; Province A is 0.125 and the distances to the positive and negative ideal solutions are 0.215, respectively. The ranking is better than that of Province B (0.317) and Province C (0.589). The ranking results are consistent with the effectiveness assessment conclusions in the market supervision reports of the three provinces in 2024, verifying the accuracy of the model.

[0043] (4) Implementation of risk tracing using fuzzy Bayesian Networks (FBNs): (a) Indicator fuzzification processing: A trapezoidal function is used to define fuzzy sets. For example, a power retention ratio of 18% belongs to "medium risk," with a membership degree of 0.8 (when 15%≤R≤20%, 0.8 = 1); a supply-demand ratio of 0.8 corresponds to "tight," with a membership degree of: ; Fuzzy processing transforms precise data into linguistic variables, adapting to the uncertainties of market risks.

[0044] (b) Network topology and CPT training: A three-layer network is constructed, with parent nodes representing market structure (supply-demand ratio, market power), price response (peak-valley response coefficient, electricity price volatility), and trading performance (green certificate turnover rate, electricity retention ratio), and child nodes representing market risk levels. Causal relationships are defined using the K2 algorithm combined with expert knowledge, such as "low supply-demand ratio → enhanced market power → price distortion." The weights of causal relationships in the topology are determined through training with 100 sets of historical risk cases to ensure the reliability of the inference.

[0045] (c) Risk reasoning and outcome analysis: Input the fuzzy membership degree of the indicators for Province B (supply-demand ratio tight μ=0.9, electricity holding capacity μ=0.8), and apply Bayes' theorem. ; The derivation yielded a high-risk probability of 91%, with the "supply and demand tension + electricity retention" path contributing 56%, triggering the risk of "market power abuse → price signal distortion." It was recommended to increase medium- and long-term contracts to 80% and limit retained electricity to ≤10%. This recommendation was adopted in the market adjustment in Province B in July 2024. After implementation, the market risk level was reduced from "high" to "medium," verifying the practicality of the model.

[0046] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0047] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0048] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A multi-dimensional method for evaluating the effectiveness of electricity market operation, characterized in that, Includes the following steps: A multi-dimensional indicator system covering the core operational dimensions of the power market is constructed, and subjective experience is combined with objective data based on a combined weighting model composed of the fusion analytic hierarchy process and the entropy weight method. Using the TOPSIS model, the distance between each province and the ideal solution is calculated based on the combined data to achieve a comprehensive ranking. After handling the uncertainty through fuzzy logic and Bayesian inference, the evaluation of the multi-dimensional power market operation effectiveness is completed.

2. The multi-dimensional electricity market operation performance evaluation method according to claim 1, characterized in that: When constructing the multi-dimensional indicator system, the system is built based on indicators of market structure, market behavior, and market performance.

3. The multi-dimensional evaluation method for the operational effectiveness of the electricity market according to claim 2, characterized in that: When obtaining market structure indicators, the market structure indicators include: Market-based electricity trading volume ratio = Medium- and long-term electricity trading volume / Total social electricity consumption × 100%; Market-based electricity price ratio = Enterprise's transaction price per kilowatt-hour / Average market price per kilowatt-hour; Medium- to long-term (annual) supply-demand ratio = sum of the maximum declared quantities of participating market entities in the annual transaction / total annual demand in the transaction market; Spot (monthly) supply-demand ratio = (Spot supply and price quotes for each period / Total maximum declared output of generating units / Net load demand); Market power is divided into wholesale market power and retail market power. It monitors the market share and structural composition of power generation companies in the market and is calculated by weighting three market power indicators: Top-m, HHI, and RSI.

4. The multi-dimensional electricity market operation performance evaluation method according to claim 3, characterized in that: When obtaining the market behavior indicators, the market behavior indicators include: Market participant participation = w1 × power generation enterprise participation + w2 × electricity sales company participation + w3 × electricity user participation, where w1 + w2 + w3 = 1 represents the different weights of the three types of market participants; Medium- and long-term transaction turnover rate = (medium- and long-term hedging transaction volume + medium- and long-term transfer transaction volume + medium- and long-term repurchase transaction volume) / medium- and long-term cumulative transaction volume × 100%; Price volatility refers to the level of fluctuation of the average market price over the most recent statistical periods. It can be calculated separately for wholesale and retail markets using the coefficient of variation. Relative Strength Index of Electricity Prices ; Peak-valley electricity price response coefficient In the formula, RC is the response coefficient, and ΔE p For the increase in peak power generation, △E v To reduce power generation during off-peak hours, S p Peak-valley price difference; Demand response volatility In the formula, For demand response volatility, The standard deviation of the response quantity This represents the average response value.

5. The multi-dimensional electricity market operation performance evaluation method according to claim 4, characterized in that: When obtaining market performance metrics, the market performance metrics include: Medium- and long-term contract signing rate = (Medium- and long-term transaction volume for different target periods / Total medium- and long-term transaction volume) × 100%; Impact of medium- and long-term transactions = Medium- and long-term contract electricity volume × Medium- and long-term contract electricity price / Settlement electricity cost × 100%; Green certificate turnover rate In the formula, TUR is the green certificate turnover rate, and Q is the green certificate turnover rate. t For green certificate trading volume, Q h Green certificate holdings; New energy utilization rate = (Total new energy power generation - Curtailed wind and solar power) / Total new energy power generation × 100%; Green electricity premium In the formula, PR is the green electricity premium rate, and P g The transaction price for green electricity, P c The benchmark price for coal-fired power generation; The peak-valley difference rate for spot market transactions = (maximum spot clearing electricity price - minimum spot clearing electricity price) / (upper limit spot clearing price - lower limit spot clearing price) × 100%, calculated based on the average daily peak-valley price difference during the statistical period; Load peak-valley difference rate = (net maximum load - net minimum load) / net maximum load × 100%, calculated based on the daily average load peak-valley difference during the statistical period, where net load = local load + externally transmitted power - externally purchased power; In the formula, Q h Electricity reserved for market entities; Q t This represents the total available electricity in the market.

6. The multi-dimensional electricity market operation performance evaluation method according to claim 5, characterized in that: When combining subjective experience with objective data, the indicators are standardized and divided into three categories: positive, negative, and interval. After eliminating the influence of dimensions using linear transformation, a judgment matrix is ​​constructed through expert questionnaires based on the analytic hierarchy process (AHP), transforming subjective experience into quantitative weights. This approach is suitable for evaluation scenarios where there is hierarchical logic among indicators, reflecting the impact of policy guidance on the importance of indicators. Furthermore, based on the entropy weight method, objective weights are assigned according to the variability of indicators, where indicators with greater data fluctuations carry more information and have higher weights.

7. The multi-dimensional electricity market operation performance evaluation method according to claim 6, characterized in that: When performing comprehensive ranking, a combined weight is formed based on the quantitative weight transformed from subjective experience and the weight result of objective weighting of indicator variability based on the entropy weight method. By constructing a decision matrix and introducing the combined weight, an ideal solution is obtained. Then, the provinces are comprehensively ranked according to the Euclidean distance between each province and the ideal solution and the closeness obtained from the Euclidean distance.

8. The multi-dimensional electricity market operation performance evaluation method according to claim 7, characterized in that: After comprehensive ranking, precise indicator values ​​are mapped to fuzzy sets based on fuzzy Bayesian networks (FBNs) to construct network topology. Causal relationships between indicators are defined through expert knowledge or data mining. Conditional probability tables are trained based on historical data. After inputting the fuzzy membership degrees of indicators, the posterior probability distribution of target nodes is derived through Bayesian formulas, enabling in-depth tracing of market risks and identification of key risk paths.

9. A multi-dimensional electricity market operation performance evaluation system, used to implement the multi-dimensional electricity market operation performance evaluation method as described in claim 1, characterized in that, include: The data acquisition and processing module is used to construct a multi-dimensional indicator system covering the core operational dimensions of the power market. It combines subjective experience with objective data based on a combined weighting model composed of the analytic hierarchy process and the entropy weight method. The evaluation module uses the TOPSIS model to calculate the distance between each province and the ideal solution based on the combined data to achieve a comprehensive ranking. After handling uncertainty through fuzzy logic and Bayesian inference, it completes the evaluation of the multi-dimensional effectiveness of the power market operation.

Images