Method and device for controlling electric energy market by virtual power plant

Abstract

The invention provides a method and device for controlling an electric energy market by a virtual power plant, and belongs to the technical field of virtual power plants, and the method comprises the steps: obtaining first data based on historical electric energy data, the first data being data for predicting the price of a future electric energy market; setting the risk preference of the virtual power plant operator based on the risk tolerance of the virtual power plant operator; and optimizing the participation strategy of the virtual power plant in the electric energy market based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator. According to the method and the device for controlling the electric energy market by the virtual power plant provided by the invention, the problem of rapid change of market conditions and internal resource states can be effectively solved.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of virtual power plants, and more specifically, relates to a method and device for controlling an electric energy market by a virtual power plant. Background Art

[0002] In the electricity market, virtual power plants, as advanced solutions for integrating distributed energy resources, face the challenge of how to participate in the market efficiently and dynamically and optimize their resource allocation. Existing technologies mostly rely on traditional forecasting methods and fixed strategy optimization frameworks, which have limitations in dealing with highly dynamic market environments and complex decision-making processes.

[0003] Traditional market price and demand forecasting methods often ignore the high nonlinearity and uncertainty of the electricity market, resulting in insufficient forecast accuracy and inability to effectively respond to rapid changes in market conditions and internal resource status. Summary of the invention

[0004] The purpose of the present disclosure is to provide a method and device for controlling the electric energy market of a virtual power plant, so as to effectively cope with the problem of rapid changes in market conditions and internal resource status.

[0005] A first aspect of an embodiment of the present disclosure provides a method for controlling an electric energy market by a virtual power plant, comprising: Obtaining first data based on historical electric energy data, wherein the first data is data for predicting future electric energy market prices; Setting the risk appetite of the virtual power plant operator based on the risk tolerance of the virtual power plant operator; The participation strategy of the virtual power plant in the electric energy market is optimized based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator.

[0006] A second aspect of the embodiments of the present disclosure provides a device for controlling an electric energy market by a virtual power plant, including: A market data analysis module, configured to obtain first data based on historical electric energy data, wherein the first data is data for predicting future electric energy market prices; A risk preference setting module, used to set the risk preference of the virtual power plant operator based on the risk tolerance of the virtual power plant operator; A strategy optimization module is used to optimize the participation strategy of the virtual power plant in the electric energy market based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator.

[0007] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above-mentioned method of controlling the electric energy market by a virtual power plant are implemented.

[0008] In a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned method of controlling the electric energy market by a virtual power plant are implemented.

[0009] The beneficial effects of a method and device for controlling the electric energy market of a virtual power plant, an electronic device, and a readable storage medium provided by the embodiments of the present disclosure are: the embodiments of the present disclosure deeply mine historical electric energy data and accurately predict future market prices, thereby providing scientific and forward-looking guidance for virtual power plants. At the same time, the embodiments of the present disclosure introduce risk preference settings for virtual power plant operators, and tailor market participation strategies according to the different risk tolerances of operators, which not only ensures stable operations, but also provides the possibility of pursuing high returns. Combining historical data, prediction results, and risk preferences, the embodiments of the present disclosure comprehensively optimize the participation strategies of virtual power plants in the electric energy market, and realize efficient dynamic adjustment of resource allocation, thereby effectively responding to rapid changes in market conditions and internal resource status. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A flowchart of a method for controlling an electric energy market by a virtual power plant provided in one embodiment of the present disclosure; Figure 2 A flowchart of a method for controlling an electric energy market by a virtual power plant provided in another embodiment of the present disclosure; Figure 3 A structural block diagram of a control device for a virtual power plant to an electric energy market provided by an embodiment of the present disclosure; Figure 4 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0012] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0013] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0014] Please refer to Figure 1 , Figure 1 A flowchart of a method for controlling an electric energy market by a virtual power plant provided in an embodiment of the present disclosure, the method comprising: S101: Obtain first data based on historical electric energy data, where the first data is data for predicting future electric energy market prices.

[0015] In this embodiment, relevant data of the electric energy market is collected and analyzed to provide data support for the generation of strategies for virtual power plants to participate in the market. The collected data may include historical clearing prices of the electric energy market, market demand, renewable energy power forecasts, weather conditions, grid operation status, etc.

[0016] For example, collecting historical clearing price data can help understand the changing trends of market prices. For example, if prices are generally higher in a specific season or time period, then in similar periods in the future, virtual power plants can prepare in advance and adjust their market participation strategies to obtain higher profits.

[0017] Accurately predicting the power output of renewable energy can help virtual power plants better plan their own power generation and energy storage strategies. For example, if the output power of wind power and photovoltaic power generation is predicted to be high in the future, the supply of electricity in the market may increase and the price may fall.

[0018] In this embodiment, the various collected data are cleaned, sorted and standardized to remove outliers and noise, ensure the quality and reliability of the data, and standardize different types of data to facilitate subsequent analysis and modeling.

[0019] In this embodiment, a variety of data analysis methods and models can be used to perform price prediction, such as time series analysis, regression analysis, machine learning algorithms, etc.

[0020] For example, time series analysis can use the time series characteristics of historical price data to predict future price trends. Regression analysis can use multiple influencing factors as independent variables to establish a mathematical relationship model between prices and these factors. Machine learning algorithms such as neural networks and support vector machines can automatically learn complex patterns in data and improve the accuracy of predictions.

[0021] The trained model is used to predict the future market price of electric energy and output the prediction results. The prediction results can be specific price values, price ranges, or trend predictions. Virtual power plants can formulate strategies for participating in the market based on these prediction data.

[0022] S102: Based on the risk tolerance of the virtual power plant operator, set the risk preference of the virtual power plant operator.

[0023] In this embodiment, risk preference refers to the operator's attitude towards the trade-off between risk and return when facing uncertainty, and reflects how much risk the operator is willing to take in the process of pursuing returns.

[0024] Risk appetite helps balance the trade-off between potential high returns and the high risks that may be faced. Risk appetite will have an impact on the quotation strategy. For example, a company with a high risk aversion may choose a more conservative quotation strategy to avoid the negative impact of large price fluctuations.

[0025] In this embodiment, risk preference can be divided into three types: risk aversion, risk neutrality and risk preference.

[0026] In this embodiment, the operator needs to collect all information related to the risk, including the analysis results of internal and external factors. Internal factors may include, but are not limited to, the operator's financial status, capital structure, profitability, market position, and long-term development strategy. For example, an operator with a sound financial status and strong profitability may have a higher tolerance for risk and be willing to take more risks in pursuit of higher returns. External factors may include market competition, changes in policies and regulations, the speed of technological progress, and changes in the macroeconomic environment. For example, in the case of intensified market competition and increased policy uncertainty, operators may reduce their tolerance for risk and adopt a more conservative strategy.

[0027] Based on the collected information, it is necessary to assess the potential risks of the operator, including the type, size, likelihood of occurrence, and possible impact of the risk. Based on the risk assessment results, the operator's risk tolerance can be determined.

[0028] In this embodiment, by evaluating the risk tolerance of the virtual power plant operator, different risk preferences are set and the bidding strategy is affected. This helps the operator to balance the potential high returns with the high risks that may be faced and develop a market participation strategy that suits its own situation.

[0029] S103: Optimizing the participation strategy of the virtual power plant in the electric energy market based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator.

[0030] In this embodiment, based on historical electricity data, including past market price fluctuations, changes in supply and demand, electricity consumption characteristics in different seasons and time periods, etc., these data can reveal the long-term trends and cyclical laws of the market. For example, if historical data shows that prices usually rise during the peak of summer electricity consumption, then when formulating participation strategies, you can consider increasing power generation capacity or adjusting the use plan of energy storage equipment in advance.

[0031] Based on the forecast of future electricity market prices in First Data, virtual power plants can have a rough judgement on future market trends. If prices are predicted to rise sharply, virtual power plants can consider reserving energy in advance or adjusting power generation plans to increase supply and obtain higher profits when prices are high. Conversely, if prices are predicted to fall, power generation can be appropriately reduced or energy storage can be increased to reduce costs.

[0032] The data is further screened and interpreted according to the risk preferences of virtual power plant operators. For conservative operators, they may prefer to rely on relatively stable trends in historical data and react relatively cautiously to future price forecasts. When formulating participation strategies, they may give priority to stable power generation and supply modes to avoid excessive risk.

[0033] In this embodiment, the participation strategies of the virtual power plant in the electric energy market may include: power generation and energy storage strategies, market transaction strategies, and risk management strategies, etc.

[0034] Based on data and risk preferences, virtual power plants can optimize the operation plan of power generation equipment. For example, when prices are high and risk preferences allow, power generation methods with higher costs but stable output, such as thermal power generation, can be increased to meet market demand and obtain higher returns. At the same time, the use strategy of energy storage equipment should be reasonably adjusted. Charge energy storage when prices are low, and discharge to supply the market when prices are high to maximize profits. Consider different trading methods, such as spot trading, futures trading, and long-term contracts. Choose a suitable trading portfolio based on risk preferences and market forecasts. Regardless of risk preferences, corresponding risk management strategies need to be formulated. This includes setting price stop loss points, diversifying investments (participating in markets in different regions or different types of energy projects), etc. For example, if prices fall to a certain level, virtual power plants can automatically start the stop loss mechanism to reduce losses. At the same time, by participating in multiple markets or projects, the impact of single market risks on the overall business is reduced.

[0035] It can be concluded from the above that this embodiment provides scientific and forward-looking guidance for virtual power plants by deeply mining historical electricity data and accurately predicting future market prices. At the same time, this embodiment introduces the risk preference setting of virtual power plant operators, and tailors market participation strategies according to the different risk tolerances of operators, which not only ensures stable operations but also provides the possibility of pursuing high returns. Combining historical data, forecast results and risk preferences, this embodiment comprehensively optimizes the participation strategy of virtual power plants in the electricity market, realizes efficient and dynamic adjustment of resource allocation, and thus effectively responds to rapid changes in market conditions and internal resource status.

[0036] In one embodiment of the present disclosure, first data is obtained based on historical electric energy data, and the first data is data for predicting future electric energy market prices, including: The historical electric energy data is input into the random forest model, and the future electric energy market price is predicted based on the random forest model to obtain the first data.

[0037] In this embodiment, a random forest model can be used for prediction. Random forest is an integrated learning method that can build multiple decision trees and output the average prediction results of the patterns of these trees to improve the performance and stability of the prediction.

[0038] In this embodiment, the random forest model is initialized and appropriate parameters are selected, including the number of decision trees, the maximum depth of the tree, the minimum number of samples required to split a node, etc. Because the number of decision trees determines the complexity and generalization ability of the model. The more the number, the more complex the model may be, but it may also better capture the complex relationships in the data; the maximum depth of the tree limits the growth of the decision tree to prevent overfitting; the minimum number of samples controls the conditions for node splitting and ensures the stability of the model.

[0039] In this embodiment, the random forest model is trained using historical data of the day-ahead market and the real-time market as training sets. The historical electric energy data contains various information related to the electric energy market price, such as market demand, supply conditions, renewable energy power, weather conditions, etc. By learning from these historical data, the random forest model can establish the relationship between the input features and the future electric energy market price. During the training process, multiple decision trees in the random forest model learn and predict the training data respectively. Each decision tree is constructed by random sampling and feature selection of the training data, which can increase the diversity and robustness of the model. Finally, the random forest model obtains a prediction of the future electric energy market price by synthesizing the prediction results of multiple decision trees.

[0040] In this embodiment, the root mean square error (RMSE) can be used to evaluate the performance of the prediction model. The calculation method of RMSE is:

[0041] Where n represents the number of samples, Expressed as actual value, Expressed as predicted value.

[0042] The root mean square error reflects the average deviation between the predicted value and the actual value. If the root mean square error is small, it means that the model's predicted results are closer to the actual value and the model's performance is good; conversely, if the root mean square error is large, it means that the model's predicted results deviate greatly from the actual value and the model's performance needs to be improved. By continuously adjusting the model's parameters and optimizing the training process, the root mean square error can be reduced and the model's prediction accuracy can be improved.

[0043] It can be concluded from the above that this embodiment first collects historical electric energy data, and then preprocesses the data to make it meet the input requirements of the random forest model. Next, the preprocessed data is input into the trained random forest model, and the model predicts the future electric energy market price based on the learned relationship, and outputs the prediction result as the first data. By using the random forest model to learn and predict historical electric energy data, the virtual power plant is provided with prediction data for the future electric energy market price, which helps the virtual power plant to formulate a more reasonable participation strategy.

[0044] In one embodiment of the present disclosure, the risk preference of the virtual power plant operator is set based on the risk tolerance of the virtual power plant operator, including: Calculate the conditional value at risk for virtual power plant operators; The risk tolerance of the virtual power plant operator is obtained based on the conditional risk value of the virtual power plant operator.

[0045] In this embodiment, the Conditional Value at Risk (CVaR) method can be introduced to consider the risk factors in the operation of the virtual power plant. CVaR is an indicator to measure risk. CVaR represents the average loss of the risk value exceeding the confidence level under a certain confidence level. By calculating CVaR, we can have a more comprehensive understanding of the degree of risk that the virtual power plant may face under different market conditions.

[0046] In one embodiment of the present disclosure, the conditional value at risk of the virtual power plant operator is calculated by a first formula; The first formula is:

[0047] in, Expressed as the average loss of risk value exceeding the confidence level α, Expressed as the value at risk at confidence level u.

[0048] In this embodiment, the virtual power plant operator's acceptance or tolerance for risk is determined based on the calculated CVaR value.

[0049] For example, the CVaR value is compared with the risk preference or risk strategy of the virtual power plant operator. If the calculated conditional risk value is low, it means that the average loss that the virtual power plant may face at a certain confidence level is small, and the operator may have a higher risk tolerance; conversely, if the conditional risk value is high, the operator's risk tolerance may be low.

[0050] From the above, it can be concluded that this embodiment can scientifically and accurately configure risk preferences for virtual power plant operators by accurately calculating the conditional risk value of virtual power plant operators and reasonably setting their risk tolerance accordingly, which helps them optimize resource allocation and achieve the best balance between risk and return while ensuring safe operation.

[0051] In one embodiment of the present disclosure, before obtaining the first data based on the historical electric energy data, the method further includes: Preprocessing the historical electric energy data to obtain second data; The first data is obtained based on the second data.

[0052] In this embodiment, historical electric energy data often has various problems, such as wrong values, missing values, and inconsistent data dimensions, which will affect the accuracy and reliability of subsequent prediction models. Therefore, before obtaining the first data for predicting the future electric energy market price based on the historical electric energy data, data preprocessing is required. By preprocessing the historical electric energy data, the data quality can be improved, laying the foundation for accurate price prediction.

[0053] In this embodiment, relevant data including historical clearing prices of the electric energy market, market demand, renewable energy power forecasts, weather conditions, grid operation status, etc. are first collected. The collected data is preliminarily checked to identify and correct obvious erroneous values. For example, data entry errors, abnormal maximum or minimum values, etc. can be corrected through manual inspection or simple logical judgment.

[0054] In this embodiment, outliers are identified by the IQR method.

[0055] Calculate the first quartile Q1 and the third quartile Q3 of the data, and then calculate the interquartile range according to the formula IQR = Q3-Q1. Define outliers as values ​​less than Q1-1.5•IQR or greater than Q3+1.5•IQR.

[0056] In this embodiment, missing values ​​are filled by linear interpolation.

[0057] For missing values ​​in the data, this embodiment can use linear interpolation to fill them. Linear interpolation inserts a reasonable value at the position of the missing value based on the trend of known data points. For example, if the electric energy data at a certain time point is missing, linear interpolation can be performed based on the data at the time points before and after it to estimate the value at that time point.

[0058] In this example, the data were normalized by the Z-score method.

[0059] In order to eliminate the dimensional differences between different data features, this embodiment can use the Z-score method to standardize the data. The formula is as follows:

[0060] Where X is the original data point, μ is the mean of the feature, and σ is the standard deviation of the feature. The standardized data has a mean of 0 and a standard deviation of 1, which is convenient for subsequent model processing and analysis.

[0061] In this embodiment, after the above preprocessing steps, high-quality second data are obtained. These data can better reflect the real situation of the electric energy market and reduce the impact of noise and abnormal values. Then, the future electric energy market price is predicted based on the second data to obtain the first data.

[0062] From the above, it can be concluded that this embodiment significantly improves the data quality by performing comprehensive pre-processing of historical electric energy data, including error value correction, outlier identification and processing, missing value filling, and data standardization, thereby ensuring the accuracy and reliability of the first data for predicting future electric energy market prices based on these data, and providing a solid foundation for strategy formulation and decision-making in the electric energy market.

[0063] In one embodiment of the present disclosure, the participation strategy of the virtual power plant in the electric energy market is optimized based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator, including: The participation strategy of virtual power plants in the electricity market is optimized based on historical electricity data, first data, risk preferences of virtual power plant operators and proximal strategies with improved adaptive trimming mechanisms.

[0064] In this embodiment, the proximal policy optimization (AT-PPO) of the improved adaptive clipping mechanism introduces a clipping mechanism for policy updates to limit the pace of policy updates. In the process of virtual power plants participating in the electric energy market, training instability caused by excessive policy updates is avoided. This makes policy updates smoother and helps improve optimization accuracy. AT-PPO has optimized exploration and utilization balance characteristics. In a complex electricity market environment, it can converge better. When participating in the market, virtual power plants need to find a balance between exploring new strategies and utilizing existing effective strategies to adapt to changing market conditions.

[0065] In this embodiment, an intelligent agent environment is constructed.

[0066] Based on historical electricity data, First Data predicts future electricity market prices and the risk preferences of virtual power plant operators. These data provide an important basis for optimizing participation strategies. Historical electricity data includes historical market clearing prices, market demand, renewable energy power forecasts, weather conditions, grid operation status, etc. First Data predicts future market prices by analyzing historical data. Risk preferences reflect the operator's trade-off between risk and return.

[0067] Based on the collected relevant data, the state space of the virtual power plant agent is defined. The state space includes market price forecasts, power demand, renewable energy forecasts, internal energy storage and load status, etc. These factors together determine the operating status and market environment of the virtual power plant at a specific moment.

[0068] The action space of the virtual power plant agent is set according to market rules. For example, the virtual power plant can choose to buy / sell electricity at each moment, declare multiple non-increasing prices when buying electricity, and declare multiple non-decreasing prices when selling electricity. The action space is the declared electricity volume and the electricity price in each segment, and the number of price segments can be set by the user.

[0069] From the above, it can be concluded that this embodiment takes into account the historical electric energy data, the first data, the risk preference of the virtual power plant operator, and the proximal strategy for improving the adaptive tailoring mechanism. The historical data and forecast data provide information on market trends and rules, the risk preference determines the decision-making tendency of the operator, and the proximal strategy is used to optimize the participation strategy of the virtual power plant.

[0070] Through the AT-PPO algorithm, combined with the above factors, the participation strategy of virtual power plants in the electric energy market is learned and optimized. During the optimization process, the algorithm continuously adjusts the purchase / sale power decisions, declared power and electricity prices of virtual power plants to achieve the goals of maximizing economic benefits and minimizing risks. At the same time, the clipping mechanism ensures the stability of strategy updates and improves the accuracy and reliability of optimization.

[0071] In an embodiment of the present disclosure, the method further includes: in response to tailoring the proximal policy of the improved adaptive tailoring mechanism, updating the proximal policy of the improved adaptive tailoring mechanism.

[0072] In this embodiment, the policy is updated by using the AT-PPO algorithm.

[0073] The objective function of the AT-PPO algorithm is to limit the range of policy updates to prevent a sharp drop in performance due to excessive updates. The calculation formula is as follows:

[0074] in, Expressed as the expected value at time t; It is expressed as the strategy probability ratio, that is, the probability ratio of the new strategy to the old strategy is , that is, the probability of taking action a in a given state With the old strategy probability The ratio of It is expressed as a clipping parameter, which is adjusted based on the deviation between the policy performance and the expected target, and can be adjusted using the following adaptive rule:

[0075] in, , Represents the clipping coefficients for the next time and current time step respectively; is the learning rate parameter, which controls the size of the adjustment step and determines the speed and sensitivity of the cropping threshold adjustment; Indicates the amount of change in performance; It is a preset performance improvement goal, that is, the level that we hope the performance will at least reach or exceed.

[0076] Implementing clipping in the objective function of the AT-PPO algorithm, using the updated Modify the objective function.

[0077] The updated objective function is:

[0078] It is expressed as an advantage function, and the calculation formula of the advantage function is as follows:

[0079] Where t is the current step, T is the total number of time steps considered or the length of the sequence, usually the length of a round; is the generalized advantage estimation (GAE) parameter used to balance the bias and variance, usually between 0 and 1; δt is the time difference error at time t, defined as: , R t Indicates that action a is taken at time t t Immediate rewards after is a discount factor used to calculate the present value of future rewards, usually between 0 and 1; V(s t ) is the state value function, which means that in state s t The expected return under t+1 ) is the next state equivalence function.

[0080] From the current strategy Sample an action, execute it and observe the reward and new state.

[0081] According to the sampled data, according to the objective function Update policy parameters.

[0082] The strategies obtained through training and optimization are applied to the market participation and resource scheduling decisions of virtual power plants.

[0083] From the above, it can be concluded that the objective function and pruning mechanism of the AT-PPO algorithm ensure the stability of strategy updates and prevent performance drops. The adaptive pruning parameters can be dynamically adjusted with the strategy performance to adapt to changes in the electricity market. Advantage function calculation provides direction for strategy updates. Parameters are sampled and updated from the current strategy, and the optimization strategy is applied to market participation and resource scheduling to improve the operational efficiency and economic benefits of virtual power plants. It enables it to better respond to market changes, enhance competitiveness, and provide strong support for virtual power plants to achieve sustainable development and efficient energy management in complex electricity markets.

[0084] In this embodiment, please refer to Figure 2 , Figure 2 A flowchart of a method for controlling the electric energy market of a virtual power plant provided in another embodiment of the present disclosure. The process mainly includes: Market data analysis: We collect a wide range of relevant data, including historical prices, market demand, renewable energy forecasts, etc., and use a variety of methods to predict market prices to provide basic information for subsequent decision-making.

[0085] Risk preference setting: Identify internal risk factors such as financial status and external risk factors such as market competition, determine the degree of risk through quantitative assessment, set different types of risk preference parameters based on the operator's ability to bear, and influence the virtual power plant's quotation and market participation strategy.

[0086] Strategy Optimization and Generation: Define the state space and action space. The state space covers factors such as market price forecasts and electricity demand, while the action space includes power purchase and sales decisions and declared prices. At the same time, design a reward function and use a proximal policy optimization algorithm to optimize the strategy.

[0087] In terms of strategy model training and updating, we use historical data for training, adjust strategy parameters based on the objective function, and update the strategy under the trimming mechanism. We regularly update the model as the market changes to ensure the adaptability of the strategy.

[0088] Policy Control and Enforcement: Apply the optimized strategy to market participation and resource scheduling decisions, and execute resource scheduling plans, such as adjusting the operation strategy of power generation equipment, adjusting energy storage equipment according to price, and selecting appropriate trading methods to reduce market risks. Through this process, virtual power plants can operate efficiently in the electric energy market, achieve reasonable allocation of resources, improve economic benefits and market competitiveness, and better adapt to the complex and changing market environment.

[0089] From the above, it can be concluded that this embodiment provides a basis for decision-making through market data analysis, and risk preference setting makes the virtual power plant quotation more reasonable. Strategy optimization and generation and model training and updating ensure strong strategy adaptability. Strategy control and execution can efficiently operate virtual power plants, reasonably allocate resources, improve economic benefits and competitiveness, enable them to better adapt to the complex and changing electricity market environment, and provide strong support for achieving sustainable development and efficient energy management.

[0090] Corresponding to the above embodiment, a method for controlling the electric energy market by a virtual power plant is provided. Figure 3 This is a block diagram of a control device for a virtual power plant to control an electric energy market according to an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 3 The control device 20 of the virtual power plant for the electric energy market includes: a market data analysis module 21, a risk preference setting module 22, and a strategy optimization module 23.

[0091] The market data analysis module 21 is used to obtain first data based on historical electric energy data, and the first data is data for predicting future electric energy market prices; A risk preference setting module 22, used to set the risk preference of the virtual power plant operator based on the risk tolerance of the virtual power plant operator; The strategy optimization module 23 is used to optimize the participation strategy of the virtual power plant in the electric energy market based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator.

[0092] In one embodiment of the present disclosure, the market data analysis module 21 is specifically used for: The historical electric energy data is input into the random forest model, and the future electric energy market price is predicted based on the random forest model to obtain the first data.

[0093] In one embodiment of the present disclosure, the risk preference setting module 22 is specifically used to: Calculate the conditional value at risk for virtual power plant operators; The risk tolerance of the virtual power plant operator is obtained based on the conditional risk value of the virtual power plant operator.

[0094] In one embodiment of the present disclosure, the risk preference setting module 22 is further configured to: The conditional value at risk of the virtual power plant operator is calculated by the first formula; The first formula is:

[0095] in, Expressed as the average loss of risk value exceeding the confidence level α, Expressed as the value at risk at confidence level u.

[0096] In one embodiment of the present disclosure, a control device 20 for a virtual power plant to an electric energy market further includes: a preprocessing module; the preprocessing module is specifically used to: Preprocessing the historical electric energy data to obtain second data; The first data is obtained based on the second data.

[0097] In one embodiment of the present disclosure, the strategy optimization module 23 is specifically used to: The participation strategy of virtual power plants in the electricity market is optimized based on historical electricity data, first data, risk preferences of virtual power plant operators and proximal strategies with improved adaptive trimming mechanisms.

[0098] In one embodiment of the present disclosure, a control device 20 for a virtual power plant to an electric energy market further includes: an update module; the update module is specifically used to: In response to tailoring the proximal policy of the improved adaptive clipping mechanism, the proximal policy of the improved adaptive clipping mechanism is updated.

[0099] See also Figure 4 , Figure 4 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 4 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules in the above-mentioned device embodiments, such as Figure 3 The functions of modules 21 to 23 are shown.

[0100] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0101] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0102] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0103] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of a method for controlling an electric energy market by a virtual power plant provided in the embodiments of the present disclosure, and can also execute the implementation methods of the electronic device described in the embodiments of the present disclosure, which will not be repeated here.

[0104] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0105] The computer-readable storage medium may be an internal storage unit of the electronic device of any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the computer-readable storage medium may also include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0106] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0107] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0108] In the several embodiments provided in the present application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0109] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0110] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0111] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A method for controlling an electric energy market by a virtual power plant, characterized in that: include: Obtaining first data based on historical electric energy data, wherein the first data is data for predicting future electric energy market prices; Setting the risk appetite of the virtual power plant operator based on the risk tolerance of the virtual power plant operator; The participation strategy of the virtual power plant in the electric energy market is optimized based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator.

2. A method for controlling the electric energy market by a virtual power plant according to claim 1, characterized in that: Obtaining first data based on historical electric energy data, wherein the first data is data for predicting future electric energy market prices, including: The historical electric energy data is input into the random forest model, and the future electric energy market price is predicted based on the random forest model to obtain the first data.

3. A method for controlling the electric energy market by a virtual power plant according to claim 1, characterized in that: The risk preferences of the virtual power plant operator are set based on the risk tolerance of the virtual power plant operator, including: Calculate the conditional value at risk for virtual power plant operators; The risk tolerance of the virtual power plant operator is obtained based on the conditional risk value of the virtual power plant operator.

4. A method for controlling the electric energy market by a virtual power plant as claimed in claim 3, characterized in that: The conditional value at risk of the virtual power plant operator is calculated by the first formula; The first formula is: in, Expressed as the average loss of risk value exceeding the confidence level α, Expressed as the value at risk at confidence level u.

5. The method for controlling the electric energy market by a virtual power plant according to claim 1, characterized in that: Before obtaining the first data based on the historical electric energy data, the method further includes: Preprocessing the historical electric energy data to obtain second data; The first data is obtained based on the second data.

6. A method for controlling the electric energy market by a virtual power plant according to claim 1, characterized in that: The optimizing the participation strategy of the virtual power plant in the electric energy market based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator includes: The participation strategy of the virtual power plant in the electric energy market is optimized based on the historical electric energy data, the first data, the risk preference of the virtual power plant operator and the proximal strategy of the improved adaptive tailoring mechanism.

7. A method for controlling the electric energy market by a virtual power plant according to claim 1, characterized in that: Also includes: In response to tailoring the proximal policy of the improved adaptive clipping mechanism, the proximal policy of the improved adaptive clipping mechanism is updated.

8. A control device for a virtual power plant to control an electric energy market, characterized in that: include: A market data analysis module, configured to obtain first data based on historical electric energy data, wherein the first data is data for predicting future electric energy market prices; A risk preference setting module, used to set the risk preference of the virtual power plant operator based on the risk tolerance of the virtual power plant operator; A strategy optimization module is used to optimize the participation strategy of the virtual power plant in the electric energy market based on the historical electric energy data, the first data and the risk preference of the virtual power plant operator.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images