A virtual power plant multi-time and space multi-scene operation reliability evaluation method and system

Abstract

A method and system for evaluating the reliability of virtual power plant operation across multiple time periods and scenarios is disclosed. The method includes: constructing a four-level evaluation index system and normalizing the indicators; calculating the subjective weights of the indicators using the analytic hierarchy process (AHP) and the objective weights using the entropy weight method, optimizing to obtain combined weights; calculating multi-time dimension scores and multi-space dimension scores based on multi-time window weights and multi-space hierarchical weights, respectively, to obtain a multi-time-space evaluation score; dynamically calculating the scenario weights of each market based on the proportion of market transaction volume and the urgency coefficient of system demand, obtaining a multi-scenario evaluation score based on the scenario weights and combined weights; identifying and quantifying uncertainties, calculating risk coefficients; and integrating the multi-time-space evaluation scores and multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, which are then output. This invention provides support for the optimized operation and market decision-making of virtual power plants.

Description

Technical Field

[0001] This invention belongs to the field of virtual power plant operation evaluation, and specifically relates to a method and system for evaluating the reliability of virtual power plant operation in multiple time and space scenarios. Background Technology

[0002] With the increasing penetration of new energy sources and the deepening of power market reforms, virtual power plants, as the core carriers aggregating distributed power sources, energy storage systems, and flexible loads, have become a key component of the new power system for flexible resource regulation. The operational reliability of virtual power plants directly affects the ability to fulfill power market transaction obligations, the safe and stable operation of the power grid, and the energy rights of users. Therefore, establishing a scientific and comprehensive reliability evaluation system is crucial.

[0003] Existing virtual power plant reliability evaluation technologies suffer from the following shortcomings: First, they have a single evaluation dimension, with traditional methods focusing on a single time dimension or a single market scenario, failing to consider the collaborative characteristics of multiple markets and making it difficult to reflect the operational status of the entire scenario. Second, they lack spatiotemporal adaptability, lacking multi-spatial-dimensional hierarchical evaluation, with fixed time weight allocation, making it impossible to dynamically adapt to the needs of different operational stages. Third, they lack uncertainty quantification, failing to fully consider the impact of factors such as fluctuations in renewable energy output, resulting in low evaluation accuracy and weak risk prediction capabilities. Fourth, their indicator system is incomplete, focusing more on technical performance while neglecting the collaborative consideration of economic benefits and user adaptability, with a single method for determining weights. Fifth, they have weak dynamic integration capabilities, lacking a dynamic weight allocation mechanism for multiple market scenarios, making it difficult to achieve accurate integration and real-time updates of evaluation results. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for evaluating the reliability of virtual power plants across multiple time and space scenarios. The aim is to enable reliability assessment and optimization decisions for virtual power plants in spot markets, medium- and long-term markets, ancillary services markets, and demand response scenarios, resolving issues such as the limited dimensionality, insufficient spatiotemporal adaptability, and lack of uncertainty quantification in existing virtual power plant reliability evaluation technologies.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution.

[0006] This invention first discloses a method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios, the method comprising the following steps: Step 1: Collect virtual power plant operation data, market transaction data, user response data, and power grid dispatch data; construct a four-level evaluation index system covering market interaction effect indicators, resource response reliability indicators, economic benefit indicators, and user adaptability indicators; and normalize the indicators. Step 2: Calculate the subjective weights of the indicators using the analytic hierarchy process (AHP), calculate the objective weights of the indicators using the entropy weight method, optimize the combined weights of the indicators by combining weighting coefficients, and allocate weights for multiple time windows and multiple spatial levels. Step 3: Calculate the multi-time dimension score and multi-space dimension score based on the multi-time window weight and multi-space level weight respectively, and obtain the multi-temporal evaluation score; dynamically calculate the scenario weight of each market based on the market transaction scale ratio and the system demand urgency coefficient, and obtain the multi-scenario evaluation score based on the scenario weight and combined weight; Step 4: Identify the uncertainties, quantify their impact using interval fuzzy theory and Monte Carlo simulation, and calculate the risk coefficient based on their impact; Step 5: Integrate the multi-temporal evaluation scores and the multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, and output them.

[0007] The present invention further includes the following preferred embodiments: The market interaction effectiveness indicators include spot market fulfillment rate, medium- and long-term contract execution rate, ancillary service response compliance rate, and demand response completion rate; the resource response reliability indicators include average response time, response volume deviation rate, and equipment failure probability; the economic benefit indicators include return on investment, net profit per unit capacity, and cost-benefit ratio; and the user adaptability indicators include the proportion of high-willing users, load elasticity matching degree, and incentive policy adjustment response rate.

[0008] The optimization of the combined weights of the indicators through the combination of weighting coefficients further includes: The Analytic Hierarchy Process (AHP) is used to construct a hierarchical structure, and a judgment matrix is ​​constructed through expert scoring to calculate the subjective weights of the indicators. w ' j ; The objective weights of indicators are calculated using the entropy weight method based on the information entropy value of the indicator data. ; The combined weights are obtained by optimization using the least squares method:

[0009] Among them, the combined weighting coefficient The value range is 0.4-0.8.

[0010] The calculation of multi-time dimension scores and multi-space dimension scores based on multi-time window weights and multi-spatial level weights, respectively, to obtain a multi-temporal and spatiotemporal evaluation score, further includes: Calculate scores across multiple time dimensions ,in and These represent the weights and evaluation scores for each time window; Calculate multi-dimensional scores ,in and These are the weights and evaluation scores for each spatial level; Introducing the temporal-spatial fusion coefficient Calculate the multi-temporal fusion score: .

[0011] The process of obtaining multi-scene evaluation scores based on the scene weights and combined weights further includes: Based on the proportion of transaction volume in each market And system requirement urgency coefficient Through formula Calculate the scenario weights for each market; For each market scenario, a scenario indicator weight vector is obtained by adjusting the combined weights. Combined with fuzzy relation matrix Calculate the contextualized evaluation score ; Calculate evaluation results for multiple market scenarios .

[0012] The allocation of multi-time window weights and multi-spatial level weights further includes: In terms of time, the evaluation period is divided into four time windows: medium-to-long-term market, spot market, ancillary services market, and demand response, with time weights assigned accordingly. ; In the spatial dimension, it is divided into three levels: resource nodes, aggregation regions, and system level, with spatial weights allocated accordingly. .

[0013] The calculation of the risk coefficient further includes: calculate:

[0014] in These represent the maximum, minimum, and average reliability scores, respectively.

[0015] This invention also discloses a virtual power plant multi-temporal-spatial-scenario operational reliability evaluation system utilizing the aforementioned virtual power plant multi-temporal-spatial-scenario operational reliability evaluation method, comprising: The indicator processing module is used to collect virtual power plant operation data, market transaction data, user response data and power grid dispatch data, and construct a four-level evaluation indicator system covering market interaction effect indicators, resource response reliability indicators, economic benefit indicators and user adaptability indicators, and normalize the indicators. The weight calculation module is used to calculate the subjective weight of indicators using the analytic hierarchy process (AHP), calculate the objective weight of indicators using the entropy weight method, optimize the combined weight of indicators by combining weighting coefficients, and allocate weights for multiple time windows and multiple spatial levels. The integrated evaluation module is used to calculate multi-time dimension scores and multi-space dimension scores based on multi-time window weights and multi-space level weights, respectively, and obtain multi-temporal and spatio-temporal evaluation scores; based on the market transaction scale ratio and system demand urgency coefficient, the scenario weights of each market are dynamically calculated, and multi-scenario evaluation scores are obtained based on the scenario weights and combined weights. The uncertainty quantification module is used to identify uncertainty factors, quantify the impact of the uncertainty factors through interval fuzzy theory and Monte Carlo simulation, and calculate the risk coefficient based on the impact. The result output optimization module is used to integrate the multi-temporal evaluation scores and the multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, and then output them.

[0016] Accordingly, this application also discloses a terminal, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps of the aforementioned virtual power plant multi-temporal and multi-scenario operation reliability evaluation method.

[0017] Accordingly, this application also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned virtual power plant multi-temporal and multi-scenario operation reliability evaluation method.

[0018] The beneficial effects of this invention are as follows: Compared with the prior art, this invention provides a method and system for evaluating the reliability of virtual power plant operation in multiple time and space scenarios. By constructing a four-dimensional indicator system of market, resources, economy, and users, it covers multiple time and space scenarios and multiple market scenarios, solving the problem of single evaluation dimensions in traditional methods. It adopts the AHP-entropy weighting method to combine weighting and dynamic time and space weight allocation, balancing subjectivity and objectivity and improving the rationality of weights. It establishes a dynamic weight allocation model for multiple market scenarios, achieving accurate adaptation to different market interaction scenarios and improving the relevance of evaluation results. It integrates interval fuzzy theory and Monte Carlo simulation to quantify the impact of multiple uncertain factors, improving the accuracy of risk assessment by more than 40%. The Pearson correlation coefficient between the evaluation results and the actual operation effect reaches 0.93, the MAE is only 2.18, and the robustness coefficient is ≥0.88, which is better than traditional single evaluation methods. It outputs differentiated optimization suggestions, providing an operable technical path for virtual power plant resource allocation, market transactions, and scheduling optimization. Attached Figure Description

[0019] Figure 1This is a flowchart of the virtual power plant multi-temporal and multi-scenario operation reliability evaluation method in this invention.

[0020] Figure 2 This is a hierarchical structure diagram of the evaluation index system in this invention; Figure 3 This is a logic diagram of multi-temporal and multi-scene fusion evaluation in this invention; Figure 4 This is a diagram of the comprehensive evaluation system architecture in this invention; Figure 5 This is a performance comparison chart of different evaluation methods in the empirical verification of this invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] The embodiments described in this application are merely some, not all, embodiments of the present invention. Based on the spirit of the present invention, other embodiments obtained by those skilled in the art without inventive effort are all within the protection scope of the present invention.

[0023] To address the shortcomings of existing technologies, this invention proposes a method and system for evaluating the reliability of virtual power plant operations across multiple time and space scenarios. This method enables multi-dimensional collaborative evaluation, improving the accuracy, dynamism, and practicality of the evaluation results. See also... Figure 1 As shown, the virtual power plant multi-temporal and multi-scenario operation reliability evaluation method disclosed in this invention includes the following steps: Step 1: Collect virtual power plant operation data, market transaction data, user response data, and power grid dispatch data, and construct a four-level evaluation index system covering market interaction effect indicators, resource response reliability indicators, economic benefit indicators, and user adaptability indicators. Normalize the indicators.

[0024] It collects virtual power plant operation data, market transaction data, user response data, and power grid dispatch data. Data sources include power trading platforms, IoT monitoring platforms, smart meters, etc. The collection cycle meets the differentiated needs from second-level (ancillary services) to monthly (medium- and long-term market).

[0025] See Figure 2The market interaction effect indicators include spot market fulfillment rate, medium and long-term contract execution rate, ancillary service response compliance rate, and demand response completion rate; the resource response reliability indicators include average response time, response volume deviation rate, and equipment failure probability; the economic benefit indicators include return on investment, net profit per unit capacity, and cost-benefit ratio; and the user adaptability indicators include the proportion of high-willing users, load elasticity matching degree, and incentive policy adjustment response rate.

[0026] The indicators are distinguished as positive or negative, and the extreme value standardization method is used for normalization to eliminate the influence of dimensions.

[0027] Step 2: Calculate the subjective weights of the indicators using the Analytic Hierarchy Process (AHP), calculate the objective weights of the indicators using the entropy weight method, optimize the combined weights of the indicators by combining weighting coefficients, and allocate weights for multiple time windows and multiple spatial levels.

[0028] (1) The subjective weights of the indicators were calculated using the Analytic Hierarchy Process (AHP). w ' j ; ① Construct a hierarchical structure: target layer (reliability of virtual power plant operation), criterion layer (4 primary indicators), and indicator layer (12 secondary indicators and 33 tertiary indicators); ② Constructing the judgment matrix: Invite 15 experts to conduct pairwise comparisons of the relative importance of indicators at the same level, and construct the judgment matrix using the 1-9 scaling method. , where a ij This indicates the degree of importance of indicator i relative to indicator j; ③ Consistency test: Calculate the largest eigenvalue of the judgment matrix. Consistency indicators Random consistency ratio (RI is the random consistency index). When CR < 0.1, the judgment matrix meets the consistency requirements. ④ Calculate subjective weights: Obtain the subjective weights by solving for the normalized eigenvectors of the judgment matrix using the eigenvector method. (j is the index number).

[0029] (2) The objective weight of the indicator is calculated based on the information entropy value of the indicator data using the entropy weight method. ; ① Data Standardization: Due to the different dimensions of the indicators, the range standardization method is used to process the indicator data. Positive indicators:

[0030] Contrarian Indicators:

[0031] in, For the original data of the i-th sample and the j-th indicator, These are the maximum and minimum values ​​of the j-th indicator, respectively; ② Calculate the entropy value of the index: (m is the number of samples), if ,but ; ③ Calculate objective weights: ,in Let be the information utility value of indicator j. The larger the information utility value, the greater the weight of the indicator.

[0032] (3) The subjective and objective weights are combined using a linear weighting method, and the combined weights are obtained by optimization using the least squares method. The combined weighting coefficients The value range is 0.4-0.8, with the optimal value being 0.6.

[0033] In terms of time, the evaluation period is divided into four time windows: medium-to-long-term market, spot market, ancillary services market, and demand response, with time weights assigned accordingly. ; In the spatial dimension, it is divided into three levels: resource nodes, aggregation regions, and system level, with spatial weights allocated accordingly. .

[0034] Step 3: Calculate the multi-time dimension score and multi-space dimension score based on the multi-time window weight and multi-space level weight respectively, and obtain the multi-temporal evaluation score; dynamically calculate the scenario weight of each market based on the market transaction scale ratio and the system demand urgency coefficient, and obtain the multi-scenario evaluation score based on the scenario weight and combined weight. Calculate scores across multiple time dimensions ,in Evaluation scores for each time window; Calculate multi-dimensional scores ,in Evaluation scores for each spatial level; Introducing the temporal-spatial fusion coefficient (Value range 0.4-0.7, optimal value 0.55), calculate the multi-temporal fusion score. .

[0035] Based on the transaction volume share of each market The contribution of virtual power plant revenue and the urgency of system demand. The scene weights are dynamically calculated using the entropy weight method. (m=1,2,3,4 correspond to the four major scenarios respectively).

[0036]

[0037] in, The percentage of the transaction volume in the m-th market (calculated by daily transaction volume for the spot market, by annual contract volume for the medium- and long-term market, by compensation amount for the ancillary services market, and by total incentive amount for demand response). The system demand urgency coefficient for the m-th market (determined by the power grid dispatch center based on factors such as system load gap and new energy fluctuations, with a value range of [0.8, 1.2]). Based on each market field scene Indicator weight vector Combined with fuzzy relation matrix Calculate the scenario-based evaluation score ; pass To achieve dynamic integration of evaluation results from multiple market scenarios, such as Figure 3 As shown.

[0038] Step 4: Identify the uncertainties, quantify their impact using interval fuzzy theory and Monte Carlo simulation, and calculate the risk coefficient based on their impact.

[0039] First, uncertainties were identified, and six key uncertainties were selected: new energy output forecasting error, spot electricity price fluctuation error, user response default rate, and equipment failure probability. The degree of influence of each uncertainty was represented by interval fuzzy numbers. For example, the influence interval of the user response default rate was [0.05, 0.25], corresponding to the fuzzy membership function.

[0040] Monte Carlo simulation was applied to generate 1000 combinations of uncertainty factors. The interval fuzzy number of each factor was substituted into a multidimensional reliability evaluation model. Through adaptation methods such as fuzzy matrix multiplication and interval operations, the interval fuzzy results of each reliability index were obtained. A weighted summation method was used to integrate the scores of each dimension index, assigning weights according to the importance of the indexes. Through weighted operations using interval fuzzy numbers, the comprehensive reliability score interval for each scenario was obtained. The upper and lower limits and midpoint values ​​of all intervals were statistically analyzed to form the overall distribution characteristics of the score intervals, including the maximum, minimum, and average values ​​of all intervals. wait.

[0041] The risk coefficient is calculated using the following formula: ,in These represent the maximum, minimum, and average reliability scores, respectively.

[0042] Low risk Medium risk. It is considered high-risk.

[0043] Step 5: Integrate the multi-temporal evaluation scores and the multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, and output them.

[0044] First, the combined weights and spatiotemporal weights are iteratively optimized based on the Particle Swarm Optimization (PSO) algorithm, as follows: (1) Initialize the particle swarm parameters, set the number of particles to 50, the number of iterations to 100, the inertia weight to 0.5, etc. Each particle corresponds to a feasible solution of a combination weight and spatiotemporal weight. The weights must satisfy the non-negativity and normalization constraints (the sum of all index weights is 1, and the sum of all spatiotemporal dimension weights is 1).

[0045] (2) With the goal of maximizing the differentiation of the comprehensive reliability score and the highest accuracy of risk level determination, an fitness function is constructed by combining the fusion effect of multi-temporal evaluation scores and multi-scenario evaluation scores. The larger the function value, the better the weight configuration.

[0046] (3) Each particle updates its velocity and position based on its own historical best position (individual optimal solution) and the historical best position of the entire particle swarm (global optimal solution), generating a new weight combination. Velocity updates need to balance inertia, individual cognition, and group cognition, while position updates need to ensure that the weights still satisfy the constraints.

[0047] (4) After each iteration, calculate the fitness value of all particles, update the individual optimal solution and the global optimal solution. If the change in the global optimal fitness value is less than the threshold (e.g., 10^-5) after 10 consecutive iterations, or the preset number of iterations is reached, stop the iteration. At this time, the global optimal solution is the optimized combined weight and spatiotemporal weight.

[0048] Secondly, robustness testing (robust coefficient when data disturbance ±15%) is combined. To verify model stability, perturbation analysis was used to test the model's robustness under data disturbances and parameter changes. Random perturbations of ±5%, ±10%, and ±15% were added to the sample data, and the coefficient of variation of the evaluation results was calculated. The fluctuation range of the evaluation results was calculated by changing the combined weighting coefficient α (range 0.4-0.8) and the time dimension weight β (range 0.4-0.7). The robustness coefficient is obtained, where This is the evaluation score after the disturbance. This indicates that the model has good robustness.

[0049] The final output includes a comprehensive reliability score, risk level, and optimization recommendations.

[0050] The above-described solution of the present invention is illustrated below through specific embodiments. Taking a 120MW virtual power plant in a certain province as the evaluation object, the virtual power plant's resource composition consists of industrial load + energy storage + distributed photovoltaic, participating in four major scenarios: spot market, medium- and long-term market, ancillary service market, and demand response, with an operating life of 3 years.

[0051] Data acquisition equipment includes: IoT monitoring terminal (sampling frequency 1 second), smart meter (sampling frequency 15 minutes), power trading platform data interface, and power grid dispatch center data transmission module; Computing equipment: Intel Core i7-12700K processor, 32GB memory, Windows Server 2019 operating system, Python 3.9 programming language, integrated NumPy, SciPy, Scikit-learn and other algorithm libraries.

[0052] See Figure 4 The specific implementation steps are as follows: A. Data Acquisition and Preprocessing The virtual power plant's operational data from January to December 2024 was collected, including market data such as daily trading volume in the spot market, execution volume of medium- and long-term contracts, number of ancillary service responses, and total demand response incentives; resource data such as renewable energy output data, load response time, and equipment failure records; economic data such as investment costs, transaction revenue, and operation and maintenance expenses; and user data such as user response willingness questionnaires and electricity consumption behavior data.

[0053] The data collection cycle meets the differentiated needs of second-level (ancillary services), minute-level (spot market), and monthly (medium- to long-term market), with data accuracy requirements of ±1% for electricity consumption, ±0.01 yuan / kWh for price, and ±0.5% for power.

[0054] Missing data were filled using linear interpolation, and outlier data were removed using the 3σ criterion, resulting in 12 complete monthly datasets.

[0055] B. Standardization of Indicators According to positive indicators Contrarian indicators The formula is normalized to obtain a standardized index matrix.

[0056] C. Weight Calculation C1. Subjective Weighting: A hierarchical structure is constructed, with experts comparing the relative importance of indicators at the same level pairwise, and a judgment matrix is ​​built using the 1-9 scaling method. Calculate the largest eigenvalue of the judgment matrix. Consistency indicators Random consistency ratio (RI is the random consistency index). When CR < 0.1, the judgment matrix meets the consistency requirement. The normalized eigenvectors of the judgment matrix are obtained by solving the eigenvector method to obtain the subjective weights. ; C2. Objective Weighting: The maximum and minimum values ​​of the indicators are obtained based on the standardized indicator matrix, using the formula... Calculate information entropy value ,in Finally passed Obtain objective weights; C3. Combination Weights: A linear weighting method is used to combine subjective and objective weights, and the proportion of subjective weights is taken. =0.6, calculate This yields the combined weight vector.

[0057] D. Multi-temporal and multi-scenario evaluation D1. Time dimension score: According to the formula According to the optimal value , , , Calculate the time dimension score ; D2. Spatial Dimension Score: According to the formula According to the optimal value , , Calculate spatial dimension score ; D3. Market Scenario Weighting: Based on the 2024 market transaction data of this virtual power plant, the spot market weight was calculated. Medium and long-term market Ancillary services market Demand Response Integration to obtain market scenario weight ; Final score: weighted by the time dimension ,calculate The virtual power plant scored 86.8 points in overall reliability in 2024.

[0058] E. Uncertainty Quantification and Risk Assessment Six factors were selected, including the fluctuation coefficient of new energy output (0.1-0.4), the fluctuation error of spot electricity price (±30%), and the user response default rate (5%-25%). 1000 simulated combinations were generated, and the score range was calculated to be [82.3, 90.5], with an average of 86.4. According to the formula... Calculate the risk coefficient It was determined to be low risk.

[0059] F. Model Validation and Optimization Comparative verification: See Figure 5 Compared with the single-level analysis method, entropy weight method, and ordinary fuzzy comprehensive evaluation method, the method of this invention has a mean absolute error (MAE) of 2.18 (reduced by 44.8%), a root mean square error (RMSE) of 2.95 (reduced by 42.4%), and a Pearson correlation coefficient of 0.93 (increased by 14.8%), indicating that the model has higher evaluation accuracy and stability. Robustness test: Add ±15% random perturbation to the sample data, and measure the robustness coefficient. ; Change the weighting coefficient of the combination (Value range 0.4-0.8), Time dimension weight (Value range 0.4-0.7), robustness coefficient This indicates that the model can maintain stable evaluation results even under conditions of data and parameter fluctuations; Optimization suggestions: Provide targeted suggestions such as "maintain the combination mode of industrial load + energy storage + distributed photovoltaic, and control the demand response incentive price at 0.7-0.9 yuan / kWh".

[0060] The virtual power plant achieved a comprehensive reliability score of 86.8 in 2024, with a low risk level. The evaluation results closely matched the actual operational performance (Pearson correlation coefficient 0.93). After implementing the optimization suggestions proposed by the method of this invention, the spot market fulfillment rate of the virtual power plant increased from 95.2% to 97.8% in the first quarter of 2025, and the return on investment increased by 3.2 percentage points, verifying the effectiveness and practicality of the invention.

[0061] The beneficial effects of this invention are as follows: Compared with the prior art, this invention provides a method and system for evaluating the reliability of virtual power plant operation in multiple time and space scenarios. By constructing a four-dimensional indicator system of market, resources, economy, and users, it covers multiple time and space scenarios and multiple market scenarios, solving the problem of single evaluation dimensions in traditional methods. It adopts the AHP-entropy weighting method to combine weighting and dynamic time and space weight allocation, balancing subjectivity and objectivity and improving the rationality of weights. It establishes a dynamic weight allocation model for multiple market scenarios, achieving accurate adaptation to different market interaction scenarios and improving the relevance of evaluation results. It integrates interval fuzzy theory and Monte Carlo simulation to quantify the impact of multiple uncertain factors, improving the accuracy of risk assessment by more than 40%. The Pearson correlation coefficient between the evaluation results and the actual operation effect reaches 0.93, the MAE is only 2.18, and the robustness coefficient is ≥0.88, which is better than traditional single evaluation methods. It outputs differentiated optimization suggestions, providing an operable technical path for virtual power plant resource allocation, market transactions, and scheduling optimization.

[0062] This invention can be a system, method, and / or computer program product. This invention also discloses a virtual power plant multi-temporal-spatial-scenario operational reliability evaluation system based on the aforementioned virtual power plant multi-temporal-spatial-scenario operational reliability evaluation method, comprising: The indicator processing module is used to collect virtual power plant operation data, market transaction data, user response data and power grid dispatch data, and construct a four-level evaluation indicator system covering market interaction effect indicators, resource response reliability indicators, economic benefit indicators and user adaptability indicators, and normalize the indicators. The weight calculation module is used to calculate the subjective weight of indicators using the analytic hierarchy process (AHP), calculate the objective weight of indicators using the entropy weight method, optimize the combined weight of indicators by combining weighting coefficients, and allocate weights for multiple time windows and multiple spatial levels. The integrated evaluation module is used to calculate multi-time dimension scores and multi-space dimension scores based on multi-time window weights and multi-space level weights, respectively, and obtain multi-temporal and spatio-temporal evaluation scores; based on the market transaction scale ratio and system demand urgency coefficient, the scenario weights of each market are dynamically calculated, and multi-scenario evaluation scores are obtained based on the scenario weights and combined weights. The uncertainty quantification module is used to identify uncertainty factors, quantify the impact of the uncertainty factors through interval fuzzy theory and Monte Carlo simulation, and calculate the risk coefficient based on the impact. The result output optimization module is used to integrate the multi-temporal evaluation scores and the multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, and then output them.

[0063] Based on the spirit of this invention, those skilled in the art will readily conceive of obtaining a computer program product based on the aforementioned method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios. The computer program product may include a computer-readable storage medium on which computer-readable program instructions are loaded to enable a processor to implement various aspects of this disclosure. That is, this application also includes a terminal comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the aforementioned method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios.

[0064] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0065] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0066] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios, characterized in that, Includes the following steps: Step 1: Collect virtual power plant operation data, market transaction data, user response data, and power grid dispatch data; construct a four-level evaluation index system covering market interaction effect indicators, resource response reliability indicators, economic benefit indicators, and user adaptability indicators; and normalize the indicators. Step 2: Calculate the subjective weights of the indicators using the analytic hierarchy process (AHP), calculate the objective weights of the indicators using the entropy weight method, optimize the combined weights of the indicators by combining weighting coefficients, and allocate weights for multiple time windows and multiple spatial levels. Step 3: Calculate the multi-time dimension score and multi-space dimension score based on the multi-time window weight and multi-space level weight respectively, and obtain the multi-temporal evaluation score; Based on the proportion of market transaction volume and the urgency coefficient of system demand, the scenario weight of each market is dynamically calculated, and the multi-scenario evaluation score is obtained based on the scenario weight and the combined weight. Step 4: Identify the uncertainties, quantify their impact using interval fuzzy theory and Monte Carlo simulation, and calculate the risk coefficient based on their impact; Step 5: Integrate the multi-temporal evaluation scores and the multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, and output them.

2. The method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios according to claim 1, characterized in that, The market interaction effectiveness indicators include spot market fulfillment rate, medium- and long-term contract execution rate, ancillary service response compliance rate, and demand response completion rate; the resource response reliability indicators include average response time, response volume deviation rate, and equipment failure probability; the economic benefit indicators include return on investment, net profit per unit capacity, and cost-benefit ratio; and the user adaptability indicators include the proportion of high-willing users, load elasticity matching degree, and incentive policy adjustment response rate.

3. The method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios according to claim 2, characterized in that, The optimization of the combined weights of the indicators through the combination of weighting coefficients further includes: The Analytic Hierarchy Process (AHP) is used to construct a hierarchical structure, and a judgment matrix is ​​constructed through expert scoring to calculate the subjective weights of the indicators. w ' j ; The objective weights of indicators are calculated using the entropy weight method based on the information entropy value of the indicator data. ; The combined weights are obtained by optimization using the least squares method: Among them, the combined weighting coefficient The value range is 0.4-0.

8.

4. The method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios according to claim 3, characterized in that, Multi-temporal and multi-spatial dimension scores are calculated based on multi-time window weights and multi-spatial level weights, respectively, to obtain a multi-temporal and spatiotemporal evaluation score, which further includes: Calculate scores across multiple time dimensions ,in and These represent the weights and evaluation scores for each time window; Calculate multi-dimensional scores ,in and These are the weights and evaluation scores for each spatial level; Introducing the temporal-spatial fusion coefficient Calculate the multi-temporal fusion score: 。 5. The method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios according to claim 4, characterized in that, The process of obtaining multi-scene evaluation scores based on the scene weights and combined weights further includes: Based on the proportion of transaction volume in each market and system requirement urgency coefficient Through formula Calculate the scenario weights for each market; For each market scenario, a scenario indicator weight vector is obtained by adjusting the combined weights. Combined with fuzzy relation matrix Calculate the contextualized evaluation score ; Calculate evaluation results for multiple market scenarios .

6. The method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios according to claim 5, characterized in that, The allocation of multi-time window weights and multi-spatial level weights further includes: In terms of time, the evaluation period is divided into four time windows: medium-to-long-term market, spot market, ancillary services market, and demand response, with time weights assigned accordingly. ; In the spatial dimension, it is divided into three levels: resource nodes, aggregation regions, and system level, with spatial weights allocated accordingly. .

7. The method for evaluating the reliability of virtual power plant operation across multiple time and space scenarios according to claim 6, characterized in that, The calculation of the risk coefficient further includes: calculate: in These represent the maximum, minimum, and average reliability scores, respectively.

8. A virtual power plant multi-temporal and multi-scenario operation reliability evaluation system, characterized in that, include: The indicator processing module is used to collect virtual power plant operation data, market transaction data, user response data and power grid dispatch data, and construct a four-level evaluation indicator system covering market interaction effect indicators, resource response reliability indicators, economic benefit indicators and user adaptability indicators, and normalize the indicators. The weight calculation module is used to calculate the subjective weight of indicators using the analytic hierarchy process (AHP), calculate the objective weight of indicators using the entropy weight method, optimize the combined weight of indicators by combining weighting coefficients, and allocate weights for multiple time windows and multiple spatial levels. The fusion evaluation module is used to calculate multi-time dimension scores and multi-space dimension scores based on multi-time window weights and multi-space level weights, respectively, and obtain multi-temporal and spatiotemporal evaluation scores. Based on the proportion of market transaction volume and the urgency coefficient of system demand, the scenario weight of each market is dynamically calculated, and the multi-scenario evaluation score is obtained based on the scenario weight and the combined weight. The uncertainty quantification module is used to identify uncertainty factors, quantify the impact of the uncertainty factors through interval fuzzy theory and Monte Carlo simulation, and calculate the risk coefficient based on the impact. The result output optimization module is used to integrate the multi-temporal evaluation scores and the multi-scenario evaluation scores to form a comprehensive reliability score, risk level, and optimization suggestions, and then output them.

9. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the virtual power plant multi-temporal and multi-scenario operation reliability evaluation method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the steps of the virtual power plant multi-temporal and multi-scenario operation reliability evaluation method according to any one of claims 1-7.

Images