Wind and light storage station group day-ahead operation plan optimization method and device and medium

Abstract

The invention relates to the technical field of energy optimization, and discloses a method and a device for optimizing a day-ahead operation plan of a wind and light storage station group, and a medium. According to the method, the target day-ahead income data and the income condition risk value are calculated by using the constructed transaction joint scene set and the probability distribution fuzzy set, and the initial wind and light storage station group day-ahead distribution robust optimization scheduling model is constructed, so that the model robustness is enhanced, and the uncertainty interference is reduced; and the operation stability and adaptability of the wind and light storage station group are enhanced. And meanwhile, the spatial-temporal correlation and multiple uncertainties of the wind and light resources are fully considered, and the adaptability of the model to the complex working conditions of the wind and light storage station group is improved. Furthermore, abstract benefit expectation and risk degree are converted into specific data indexes, the specific data indexes are fused into an initial wind and light storage station group day-ahead distribution robust optimization scheduling model, fine quantification of decision making is realized, and station group operation decision is guaranteed to consider economic benefits and risk prevention and control.

Description

Technical Field

[0001] The present invention relates to the field of energy optimization technology, and in particular to a method, device and medium for optimizing the day-ahead operation plan of a wind-solar-storage station group. Background Art

[0002] In the process of vigorously promoting energy marketization reform, it has become an inevitable trend for large-scale renewable energy sites to participate in market competition. At present, large-scale renewable energy sites are paying more and more attention to how to improve economic benefits and energy efficiency. However, current technologies are mostly used for the optimization of day-ahead operation plans of single renewable energy sites, and there is little research on the optimization of day-ahead operation plans of large-scale sites. In addition, when dealing with the uncertainty of new energy, stochastic optimization or robust optimization is generally applied, and the consideration of return risk is less. Summary of the invention

[0003] In view of this, the present invention provides a method, device and medium for optimizing the day-ahead operation plan of a wind, solar and storage station group, so as to solve the problem that the current technology is mostly applied to the day-ahead operation plan optimization of a single renewable energy station, there is less research on the day-ahead operation plan optimization of large-scale station groups, and when dealing with the uncertainty of new energy, random optimization or robust optimization is generally applied, and there is less consideration of return risk.

[0004] In a first aspect, the present invention provides a method for optimizing a day-ahead operation plan of a wind-solar-storage station group, the method comprising:

[0005] Obtain the transaction data set of the wind, solar and storage farm clusters participating in the day-ahead energy market and the historical predicted power error data set of the wind, solar and storage farm clusters; establish a transaction joint scenario set and a probability distribution fuzzy set based on the transaction data set and the historical predicted power error data set, and the probability distribution fuzzy set is used to characterize the probability distribution of the day-ahead short-term predicted active power errors of wind and light considering the spatiotemporal correlation of wind and solar resources; calculate the target day-ahead revenue data and the return conditional risk value of the wind, solar and storage farm clusters based on the transaction joint scenario set and the probability distribution fuzzy set; establish the initial day-ahead distributed blue-rod optimization scheduling model of the wind, solar and storage farm clusters based on the transaction joint scenario set, the historical predicted power error data set, the target day-ahead revenue data and the return conditional risk value; solve the initial day-ahead distributed blue-rod optimization scheduling model of the wind, solar and storage farm clusters to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm clusters.

[0006] The method for optimizing the day-ahead operation plan of a wind-solar-storage station group provided by the present invention can truly reflect the actual transaction history and past prediction error status of the wind-solar-storage station group by obtaining the transaction data set of the wind-solar-storage station group participating in the day-ahead energy market and the historical prediction power error data set, and can further ensure that the established transaction joint scenario set and probability distribution fuzzy set can better fit the actual operation scenario of the wind-solar-storage station group. Furthermore, the constructed transaction joint scenario set and probability distribution fuzzy set are used to calculate the target day-ahead revenue data and the risk value of the revenue condition and construct the initial day-ahead distributed robust optimization scheduling model of the wind-solar-storage station group. The probability distribution fuzzy set is used to carefully characterize the probability distribution of the short-term prediction of active power error of the wind-solar-storage station group, which strengthens the robustness of the model, reduces uncertainty interference, and enhances the operational stability and adaptability of the wind-solar-storage station group. At the same time, the spatiotemporal correlation and multiple uncertainties of wind and solar resources are fully considered, and the model's adaptability to the complex working conditions of the wind-solar-storage station group is improved. Furthermore, the abstract profit expectations and risk levels are converted into specific data indicators and integrated into the initial day-ahead distributed optimization scheduling model of the wind, solar and storage station cluster, which enables precise quantification of decision-making and can weigh the profit acquisition and risk response strategies under different working conditions, thereby ensuring that the operation decisions of the station cluster take into account both economic benefits and risk prevention and control.

[0007] In an optional implementation, obtaining a transaction data set and a historical prediction power error data set of a wind, solar and storage station group participating in a day-ahead energy market includes:

[0008] Obtain the historical cleared electricity quantity data and historical cleared electricity price data of wind, solar and storage farm clusters participating in the day-ahead energy market, the historical data of the day-ahead short-term predicted active power of wind and solar power of the wind, solar and storage farm clusters, and the real-time historical data of the maximum active power that can be generated by wind and solar power; determine the transaction data set based on the historical cleared electricity quantity data and historical cleared electricity price data; determine the historical predicted power error data set based on the historical data and real-time historical data.

[0009] The method for optimizing the day-ahead operation plan of a wind-solar-storage farm cluster provided by the present invention determines the transaction data set and the historical prediction power error data set by obtaining the historical clearing electricity volume data and historical clearing electricity price data of the wind-solar-storage farm cluster participating in the day-ahead energy market, the historical data of the day-ahead short-term predicted active power of the wind-solar-storage farm cluster, and the real-time historical data of the maximum wind-solar-storage active power that can be generated. It can accurately characterize the operation characteristics, market adaptability and performance of the wind-solar-storage farm cluster in the historical stage when facing wind and solar uncertainties, and provide support for the subsequent optimization of the day-ahead operation plan of the wind-solar-storage farm cluster.

[0010] In an optional implementation, based on the transaction data set and the historical prediction power error data set, a transaction joint scenario set and a probability distribution fuzzy set are established, including:

[0011] Based on the transaction data set, a transaction joint scenario set is generated through backward reduction method. Based on the historical predicted power error data set, a predicted power error support set is established. Based on the predicted power error support set, Copula theory and Wasserstein distance are used to establish a probability distribution fuzzy set.

[0012] The method for optimizing the day-ahead operation plan of a wind-solar-storage-field station group provided by the present invention generates a trading joint scenario set by processing the trading data set with the backward reduction method, which can reasonably simplify and integrate complex trading situations on the basis of retaining the key features and information of the trading data. Furthermore, a predicted power error support set is established based on the historical predicted power error data set, which can centrally summarize and organize the errors between the predicted power and the actual power of wind-solar power generation in the past, provide the original material for the subsequent construction of the probability distribution fuzzy set from the data level, and determine the basic range and boundary conditions for the subsequent characterization of the probability distribution. Furthermore, when constructing the probability distribution fuzzy set, the Copula theory can well capture the dependency structure between the variables, and then integrate multiple related factors to accurately describe their joint probability distribution characteristics. At the same time, the Wasserstein distance can measure the differences between different probability distributions, and then define the range of the probability distribution fuzzy set, so that the constructed fuzzy set can not only cover a variety of possible probability distribution situations, but also reasonably limit the range to avoid being too broad and losing practical application value, and finally enable the obtained probability distribution fuzzy set to comprehensively and accurately characterize the probability distribution uncertainty of the wind-solar predicted power error.

[0013] In an optional implementation, based on the transaction joint scenario set and the probability distribution fuzzy set, the target day-ahead revenue data and the revenue conditional risk value of the wind, solar and storage station group are calculated, including:

[0014] Based on the transaction joint scenario set, the initial day-ahead profit data of the wind, solar and storage farm cluster is calculated; based on the initial day-ahead profit data, the transaction joint scenario set and the probability distribution fuzzy set, the target day-ahead profit data of the wind, solar and storage farm cluster is calculated; based on the target day-ahead profit data, the profit conditional risk value is calculated.

[0015] The method for optimizing the day-ahead operation plan of a wind, solar, and storage station group provided by the present invention calculates the initial day-ahead revenue data through a transaction joint scenario set, and can quantify the revenue that the station group can theoretically obtain in each scenario based on different typical transaction situations. Furthermore, on the basis of the initial day-ahead revenue data, the target day-ahead revenue data is calculated in combination with the transaction joint scenario set and the probability distribution fuzzy set, taking into account the uncertainty brought by the wind and solar power generation forecast power error and a variety of possible probability distribution situations, thereby achieving a more accurate assessment of the revenue of the wind, solar, and storage station group. Finally, by calculating the risk value of the revenue condition, the risks faced by the wind, solar, and storage station group can be quantified, providing a key scientific basis for operators when making decisions, so that they can more reasonably plan the operation strategy, market transaction strategy, etc. of the station group according to their risk tolerance and revenue goals, and effectively balance the relationship between revenue acquisition and risk control.

[0016] In an optional implementation, based on the transaction joint scenario set, the historical prediction power error data set, the target day-ahead revenue data and the revenue condition risk value, an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage station cluster is established, including:

[0017] Based on the historical forecast power error data set, target day-ahead yield data and yield condition risk value, the optimization target is determined; based on the preset constraint set, transaction joint scenario set and optimization target, the initial day-ahead distributed blue-rod optimization scheduling model of the wind, solar and storage station cluster is established.

[0018] The method for optimizing the day-ahead operation plan of a wind, solar, and storage station group provided by the present invention determines the optimization target by combining the target day-ahead revenue data and the risk value of the revenue condition, breaking the traditional single-dimensional decision-making mode that only focuses on maximizing revenue, comprehensively weighing the revenue acquisition and the risk conditions faced during the operation of the wind, solar, and storage station group, and fully recognizing the impact of the volatility that the uncertainty of wind and solar power generation may bring to the revenue, so that the optimization target is more in line with the actual operation needs. Furthermore, an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar, and storage station group is established based on the preset constraint condition set, the transaction joint scenario set, and the determined optimization target. By integrating the actual operation constraints such as the wind, light, and storage operation constraints of the wind, solar, and storage station group and the day-ahead operation plan constraints of the entire station into the model, it is ensured that the operation plan generated by the constructed model is feasible in actual operation, and at the same time, it can fully optimize the resource allocation and scheduling strategy under the established goals, thereby improving the overall operation efficiency and stability.

[0019] In an optional implementation, the day-ahead distributed blue-rod optimization scheduling model of the initial wind-solar-storage farm group is solved to obtain the day-ahead operation plan optimization result of the wind-solar-storage farm group, including:

[0020] By using linear decision rules and duality theory, the initial day-ahead blue-rod optimization scheduling model of the wind, solar and storage farm group is reconstructed to obtain the day-ahead blue-rod optimization scheduling model of the target wind, solar and storage farm group; the day-ahead blue-rod optimization scheduling model of the target wind, solar and storage farm group is solved to obtain the optimization result of the day-ahead operation plan of the wind, solar and storage farm group.

[0021] The method for optimizing the day-ahead operation plan of a wind-solar-storage farm cluster provided by the present invention utilizes linear decision rules and duality theory to reconstruct the model, thereby improving the model structure. This makes the day-ahead distributed robust optimization scheduling model of the reconstructed target wind-solar-storage farm cluster easier to solve and can better reflect the relationship between the variables and the balance between the optimization objectives and the constraints, thereby improving the solution efficiency and accuracy.

[0022] In a second aspect, the present invention provides a device for optimizing the day-ahead operation plan of a wind-solar-storage station group, the device comprising:

[0023] An acquisition module is used to obtain the transaction data set of wind, solar and storage farms participating in the day-ahead energy market and the historical predicted power error data set of the wind, solar and storage farms; a first establishment module is used to establish a transaction joint scenario set and a probability distribution fuzzy set based on the transaction data set and the historical predicted power error data set, and the probability distribution fuzzy set is used to characterize the probability distribution of the day-ahead short-term predicted active power errors of wind and light considering the spatiotemporal correlation of wind and solar resources; a calculation module is used to calculate the target day-ahead revenue data and the return conditional risk value of the wind, solar and storage farms based on the transaction joint scenario set and the probability distribution fuzzy set; a second establishment module is used to establish an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage farms based on the transaction joint scenario set, the historical predicted power error data set, the target day-ahead revenue data and the return conditional risk value; a solution module is used to solve the initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage farms to obtain the day-ahead operation plan optimization result of the wind, solar and storage farms.

[0024] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0025] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for optimizing the day-ahead operation plan of a wind, solar and storage station group according to the first aspect or any corresponding embodiment thereof.

[0026] In a fifth aspect, the present invention provides a computer program product, including computer instructions, which are used to enable a computer to execute the method for optimizing the day-ahead operation plan of a wind, solar and storage station group according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 It is a flow chart of a method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention;

[0029] Figure 2 is a flow chart of another method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention;

[0030] Figure 3 It is a flow chart of another method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention;

[0031] Figure 4 It is a flow chart of a method for optimizing the day-ahead operation plan of a large-scale wind-solar-storage station group considering deviation assessment penalties and distributed robust conditional risk value according to an embodiment of the present invention;

[0032] Figure 5 is a structural block diagram of a device for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention;

[0033] Figure 6 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0035] Most existing technologies are aimed at the day-ahead operation optimization of a single renewable energy site or a single "renewable energy + energy storage" site, and there is little research on the day-ahead operation plan optimization of large-scale wind, solar and energy storage sites. In addition, when dealing with the uncertainty of new energy, random optimization, such as the scenario method, is generally applied, which can only guarantee the economy of the optimization results, but not the robustness, and rarely considers the spatiotemporal correlation of wind and solar resources, and the consideration of multiple uncertainties is also insufficient. In addition, when optimizing the day-ahead operation plan, current technologies generally do not consider the risks of short-term wind and solar forecast errors in day-ahead returns, and generally do not allow over-generation within the scope of the assessment range.

[0036] According to an embodiment of the present invention, an embodiment of a method for optimizing the day-ahead operation plan of a group of wind, solar and storage stations is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0037] In this embodiment, a method for optimizing the day-ahead operation plan of a wind-solar-storage station group is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers, etc. Figure 1 is a flow chart of a method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0038] Step S101, obtaining a transaction data set of a wind, solar, and storage station group participating in a day-ahead energy market and a historical prediction power error data set of the wind, solar, and storage station group.

[0039] Specifically, the transaction data set is used to reflect the transaction status of wind, solar and storage station clusters participating in the day-ahead energy market, which can be obtained through the power market trading platform and may include historical data on the cleared electricity volume and cleared electricity price of wind, solar and storage station clusters participating in the day-ahead energy market.

[0040] Furthermore, the historical prediction power error data set of the wind, solar and storage station group is used to reflect the error of the short-term predicted active power of wind and solar power in the past period of time, which can be obtained and determined through the new energy power prediction platform.

[0041] Step S102, establishing a transaction joint scenario set and a probability distribution fuzzy set based on the transaction data set and the historical prediction power error data set.

[0042] Among them, the transaction joint scenario set represents a combination of different situations constructed by comprehensively considering various factors that affect the wind, solar and storage station groups to conduct electricity trading and other related activities. It can cover different electricity market price fluctuations, the actual output level of wind and solar resources (wind speed conditions of wind power, light intensity of photovoltaic power generation, etc.), energy storage charging and discharging strategies, and electricity load demand and other scenario elements, combining these elements to form specific and analyzable scenarios.

[0043] In this embodiment, the transaction joint scenario set represents a typical joint scenario of wind, solar and storage station groups participating in the clearing electricity volume and clearing electricity price in the day-ahead energy market.

[0044] Furthermore, probability distribution fuzzy sets are used to characterize the probability distribution of the day-ahead short-term forecast active power error of wind and solar power taking into account the spatiotemporal correlation of wind and solar power resources.

[0045] Specifically, by establishing a joint transaction scenario set, it is possible to extract representative and typical transaction situations from massive and complex historical transaction data, simplify and focus the analysis process, and thus better grasp the different market environments and transaction conditions faced by wind, solar and storage station clusters in the day-ahead energy market.

[0046] Furthermore, due to the uncertainty of wind and solar power generation, its predicted power error does not follow a fixed and clear probability distribution, but has multiple possibilities and is affected by multiple factors such as time and space. Therefore, by establishing a probability distribution fuzzy set, this complex and fuzzy uncertainty can be quantitatively described, so that the risks brought by the uncertainty of wind and solar power generation can be fully considered in subsequent calculations and model construction, so that the operation plan of wind and solar storage stations can be formulated more scientifically and reasonably, thereby enhancing the ability to deal with uncertainty.

[0047] Step S103, based on the transaction joint scenario set and the probability distribution fuzzy set, calculate the target day-ahead revenue data and the revenue conditional risk value of the wind, solar and storage station group.

[0048] Specifically, the target day-ahead revenue data is the revenue of the wind, solar and storage station cluster based on the typical joint scenario of the day-ahead cleared electricity volume and cleared electricity price, as well as the probability distribution fuzzy set of the day-ahead short-term forecast active power error of wind and solar.

[0049] Furthermore, the return conditional value at risk (CVaR) is an indicator that measures the average degree of loss that a wind, solar and storage station cluster may suffer when the return is lower than a certain threshold under a certain confidence level. It can quantify the potential risks faced during the operation of the station cluster.

[0050] Step S104, based on the transaction joint scenario set, the historical prediction power error data set, the target day-ahead revenue data and the revenue condition risk value, establish an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage station cluster.

[0051] Specifically, on the basis of considering the typical joint scenarios of day-ahead cleared electricity quantity and cleared electricity price, the model is constructed by combining the historical forecast power error data set, the day-ahead cleared electricity quantity and the typical joint scenarios of cleared electricity price, which strengthens the robustness of the model, reduces the interference of uncertainty, and enhances the operational stability and adaptability of the wind, solar and storage station group. At the same time, the spatiotemporal correlation and multiple uncertainties of wind and solar resources are fully considered, and the adaptability of the model to the complex working conditions of the wind, solar and storage station group is improved. Furthermore, the abstract benefit expectations and risk levels are transformed into specific data indicators and integrated into the initial day-ahead distributed robust optimization scheduling model of the wind, solar and storage station group, which realizes the precise quantification of decision-making, can balance the benefit acquisition and risk response strategies under different working conditions, and thus can ensure that the operational decisions of the station group take into account both economic benefits and risk prevention and control.

[0052] Step S105, solving the day-ahead distributed blue-rod optimization scheduling model of the initial wind, solar and storage farm cluster to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm cluster.

[0053] Specifically, the solver can be used to solve the constructed initial day-ahead distributed blue-robust optimization scheduling model of the wind, solar and storage farm cluster and output the day-ahead operation plan optimization results of the wind, solar and storage farm cluster.

[0054] The method for optimizing the day-ahead operation plan of the wind, solar and storage farm cluster provided in this embodiment can truly reflect the actual transaction history and past prediction error status of the wind, solar and storage farm cluster by obtaining the transaction data set of the wind, solar and storage farm cluster participating in the day-ahead energy market and the historical prediction power error data set, and can further ensure that the established transaction joint scenario set and probability distribution fuzzy set can better fit the actual operation scenario of the wind, solar and storage farm cluster. Furthermore, the constructed transaction joint scenario set and probability distribution fuzzy set are used to calculate the target day-ahead revenue data and the risk value of the revenue condition and construct the initial day-ahead distributed robust optimization scheduling model of the wind, solar and storage farm cluster. The probability distribution fuzzy set is used to carefully characterize the probability distribution of the short-term prediction of the active power error of the wind and solar day-ahead, which strengthens the robustness of the model, reduces the interference of uncertainty, and enhances the operational stability and adaptability of the wind, solar and storage farm cluster. At the same time, the spatiotemporal correlation and multiple uncertainties of wind and solar resources are fully considered, and the adaptability of the model to the complex working conditions of the wind, solar and storage farm cluster is improved. Furthermore, the abstract profit expectations and risk levels are converted into specific data indicators and integrated into the initial day-ahead distributed optimization scheduling model of the wind, solar and storage station cluster, which enables precise quantification of decision-making and can weigh the profit acquisition and risk response strategies under different working conditions, thereby ensuring that the operation decisions of the station cluster take into account both economic benefits and risk prevention and control.

[0055] In this embodiment, a method for optimizing the day-ahead operation plan of a wind-solar-storage station group is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers, etc. Figure 2 is a flow chart of a method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0056] Step S201, obtaining a transaction data set of the wind, solar and storage farm cluster participating in the day-ahead energy market and a historical prediction power error data set of the wind, solar and storage farm cluster.

[0057] Specifically, the above step S201 includes:

[0058] Step S2011, obtain the historical clearing electricity data and historical clearing electricity price data of the wind, solar and storage station group participating in the day-ahead energy market, the historical data of the wind and solar day-ahead short-term predicted active power of the wind, solar and storage station group, and the real-time historical data of the maximum active power that can be generated by the wind and solar.

[0059] Specifically, the historical clearing electricity data indicates the actual scale of electricity transactions in the past day-ahead energy market transactions of wind, solar and storage farms. It can reflect the electricity supply level and market share change trend of the farm group under different market demands, different seasons, and different market competition environments. It is one of the key indicators for measuring the market transaction activity and scale of the farm group, and can be obtained through the electricity market trading platform.

[0060] Furthermore, the clearing electricity price represents the price per unit of electricity determined in the market clearing process by a combination of factors such as market supply and demand, power generation costs, and quotations from other power generation entities. The historical clearing electricity price data represents the price of each transaction in the past when the wind, solar, and storage station cluster participated in the day-ahead energy market, which can be obtained through the new energy power forecasting platform.

[0061] Furthermore, the historical data of the short-term predicted active power of wind and solar power in the wind and solar storage station group can reflect the accuracy and reliability of the prediction model in different time periods and weather conditions, and can be obtained through the new energy power prediction platform.

[0062] Furthermore, the real-time historical data of the maximum active power that can be generated by wind and solar power is used to reflect the maximum power that can actually be achieved by wind farms and photovoltaic power stations in the corresponding period, that is, the maximum active power value that can be generated theoretically under the actual natural conditions at that time (actual wind speed, light intensity, etc.), which can be obtained through the actual monitoring system or related records. For example, at a certain moment in a wind farm, the actual wind speed makes the maximum power generated by the wind turbine reach 120 megawatts, so the actual maximum active power data at different times and dates are summarized as real-time historical data.

[0063] Step S2012: determining a transaction data set based on historical clearing electricity quantity data and historical clearing electricity price data.

[0064] Specifically, according to the description of step S101, the acquired historical clearing electricity quantity data and historical clearing electricity price data are integrated to form a corresponding transaction data set.

[0065] Step S2013: determining a historical prediction power error data set based on the historical data and the real-time historical data.

[0066] Specifically, the historical data of the short-term predicted active power of wind and solar power in the wind and solar storage station group can be subtracted from the real-time historical data of the maximum active power that can be generated by wind and solar power to obtain the historical data of the error of the short-term predicted active power of wind and solar power in the day before, that is, the historical predicted power error data set.

[0067] In an optional implementation, the historical data of the short-term predicted active power of the wind and solar power storage station group before the day can be matched one by one with the real-time historical data of the maximum active power that can be generated by the wind and solar power, and then the difference calculations are performed one by one according to the power error calculation formula (power error = short-term predicted active power of the wind and solar power before the day - maximum active power that can be generated by the wind and solar power) to obtain the corresponding power error values ​​each time, and these power error values ​​are sorted in chronological order or other reasonable ways to form the corresponding historical predicted power error data set.

[0068] Step S202, based on the transaction data set and the historical prediction power error data set, establish a transaction joint scenario set and a probability distribution fuzzy set.

[0069] Specifically, the above step S202 includes:

[0070] Step S2021, based on the transaction data set, a transaction joint scenario set is generated through backward reduction processing.

[0071] Among them, the Backward Reduction Method refers to a technology used to reduce the number of scenarios in power market analysis and operation. It can gradually eliminate those scenarios with less impact on the system or similar impact from the original large number of scenarios, and retain representative and important scenarios.

[0072] Specifically, based on the historical data of the cleared electricity volume and cleared electricity price of the day-ahead energy market of the power market trading platform, the cleared electricity price of a large number of wind, solar and storage station clusters can be obtained. Combined scenario with power clearing These joint scenarios can reflect the possible combinations of clearing electricity prices and clearing electricity quantities of wind, solar and storage stations under different market conditions.

[0073] Furthermore, the obtained joint scenes can be processed by backward reduction method, and N s The number of representative scenes is N. s .

[0074] Furthermore, each reduced scene s is assigned a probability π s , used to represent the possibility of each scenario occurring in the actual market.

[0075] Step S2022: establishing a predicted power error support set based on the historical predicted power error data set.

[0076] Specifically, the vector of uncertain variables of wind and light forecast errors is It is composed of the order of the short-term predicted active power errors of wind and light in the dispatching period of each wind-solar-storage station in the wind-solar-storage station group. For example, the short-term predicted active power error of wind in station No. 1 during the dispatching period is: The optical short-term predicted active power error of station 1 during the dispatch period Wind short-term forecast active power error at station 2 during the dispatch period The optical short-term predicted active power error of station 2 during the dispatch period ...、N in the scheduling period WSS Active power error of wind short-term forecast at station No. During the scheduling period N wSs Optical short-term prediction active power error of station No. As shown in the following relationship (1):

[0077]

[0078] Where: represents the historical prediction power error data set; N Wss It indicates the number of field stations in the wind-solar-storage station group; T indicates the total dispatching period.

[0079] Furthermore, the historical samples are transformed into Standardization, standardized uncertain variables The support set Ω can be defined as the following relation (2):

[0080]

[0081] Where: represents the tth element; q represents The bounds of all elements in under the support set Ω.

[0082] Further, assuming The empirical distribution, true distribution and distribution fuzzy set of are q can be obtained by solving the following minimization problem using a nested binary search method, as shown in equation (3):

[0083]

[0084] In the formula: ζ represents a relatively high confidence level.

[0085] Furthermore, after acquisition, the compact form of the support set driven by historical data, i.e., the predicted power error support set θ, can be expressed as the following relation (4):

[0086]

[0087] Where: A = [I; -I], which means a matrix consisting of the unit matrix I and its negative matrix; Represents the mean vector of the boundary q, the all-one vector e and δ The matrix formed.

[0088] Step S2023, based on the predicted power error support set, a probability distribution fuzzy set is established using Copula theory and Wasserstein distance.

[0089] Specifically, when obtaining the uncertain variable vector After the support set Θ of the wind-solar-storage station group, considering that the stations within the wind-solar-storage station group are in close geographical locations, the wind-solar-storage station's wind and solar resources have a certain correlation in time and space. The Gaussian function is used as the kernel function of the non-parametric kernel density estimation, and based on the historical data of the short-term prediction active power error of wind and solar, the probability density function of the short-term prediction error of wind and solar in period t is obtained. Then we can get the corresponding marginal distribution function of time period t:

[0090] Furthermore, according to Sklar's theorem, the Copula function is used to describe the spatiotemporal correlation between random variables and can connect the respective marginal distribution functions into a joint distribution function.

[0091] In this embodiment, Frank-Copula is selected as the Copula function, the parameter θ is obtained by the maximum likelihood method, and the mean of the joint distribution of multiple periods is used as An estimate of the true distribution of As shown in the following relation (5):

[0092]

[0093] Furthermore, based on the above, the fuzzy set of uncertain variables based on Wasserstein distance It can be expressed as the following relation (6):

[0094]

[0095] Where: The value of the inf problem on the left side of the inequality sign is Potential distribution of Φ(Θ) is the set of all possible distributions in the support set) and the empirical distribution The Wasserstein distance of ; Π is the joint distribution of the latent distribution and the empirical distribution; represents an arbitrary norm; the right side of the inequality sign represents the radius of the Wasserstein sphere; β represents the confidence of the Wasserstein sphere; Represents the sample mean.

[0096] Step S203, based on the transaction joint scenario set and the probability distribution fuzzy set, calculate the target day-ahead revenue data and the revenue conditional risk value of the wind, solar and storage station group.

[0097] Specifically, the above step S203 includes:

[0098] Step S2031, based on the transaction joint scenario set, calculate the initial day-ahead revenue data of the wind, solar and storage station group.

[0099] Specifically, combining the transaction joint scenario set, we can get the day-ahead profit of the s-th scenario: s , as shown in the following relation (7):

[0100]

[0101] Where: represents the day-ahead return at time t for the s-th scenario, which can be calculated using the following equation (8): The penalty cost for the deviation power exceeding the assessment margin at time t in the sth scenario can be calculated by the following equation (9): The energy storage charging and discharging cost at time t in the sth scenario can be calculated by the following equation (10): The penalty cost of wind and solar power abandonment at time t in the sth scenario can be calculated using the following equation (11):

[0102]

[0103] In the formula, represents the clearing electricity price of the wind, solar and storage station group at time t in the sth scenario; represents the planned power generation of the day-ahead operation at the tth moment in the sth scenario; π assess It indicates the deviation assessment coefficient; represents the deviation assessment power of the wind-solar-storage station group at the tth moment in the sth scene; c krepresents the lifetime degradation cost of the energy storage system at the kth station; represents the day-ahead planned charging / discharging power of the energy storage system at the t-th moment of the s-th scenario; represents the charging / discharging efficiency of the energy storage system at the kth station; Cur represents the unit penalty cost for wind and solar power abandonment; represents the short-term predicted wind power / PV power of the kth station at time t; They represent the day-ahead planned power generation of wind power / photovoltaic power at the k-th site in the s-th scenario at time t; They represent the short-term wind power / photovoltaic prediction errors of the kth station at the tth moment respectively.

[0104] Step S2032, based on the initial day-ahead revenue data, the transaction joint scenario set and the probability distribution fuzzy set, calculate the target day-ahead revenue data of the wind, solar and storage station group.

[0105] Specifically, the initial day-ahead profit data profit is calculated s On this basis, considering the typical joint scenarios of day-ahead cleared electricity quantity and cleared electricity price as well as the probability distribution fuzzy set of the day-ahead short-term forecast active power error of wind and solar power, the day-ahead revenue of the wind-solar-storage station group, i.e. the target day-ahead revenue data, can be expressed as

[0106] Step S2033, calculate the return conditional value at risk based on the target day return data.

[0107] Specifically, the risk value of return is the minimum return under a certain confidence level, and the conditional risk value of return is the expected value of the return less than the risk value under the confidence level, as shown in the following relationship (12):

[0108]

[0109] In the formula: σ represents the conditional risk value of the previous day's return, i.e., the conditional risk value of the return; ξ represents the risk value; α represents the confidence level of the conditional risk value; Represents the maximum value between 0 and .

[0110] Step S204: Based on the transaction joint scenario set, the historical forecast power error data set, the target day-ahead revenue data and the risk value of revenue conditions, an initial wind, solar and storage station cluster day-ahead distributed optimization scheduling model is established. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0111] Step S205, solve the day-ahead optimization dispatch model of the initial wind, solar and storage farm group to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm group. Figure 1Step S105 of the illustrated embodiment will not be described in detail here.

[0112] The method for optimizing the day-ahead operation plan of a wind, solar and storage farm group provided in this embodiment can accurately characterize the operation characteristics, market adaptability and performance of the wind, solar and storage farm group in the historical stage when facing the uncertainty of wind and solar power by obtaining the transaction data set of the wind, solar and storage farm group participating in the day-ahead energy market and the historical predicted power error data set. Furthermore, by using the backward reduction method to process the transaction data set to generate a transaction joint scenario set, it is possible to reasonably simplify and integrate complex transaction situations on the basis of retaining the key features and information of the transaction data. Furthermore, based on the historical predicted power error data set, a predicted power error support set is established, which can centrally summarize and organize the errors between the predicted power and the actual power of wind and solar power generation in the past, and provide the original material for the subsequent construction of probability distribution fuzzy sets from the data level, and determine the basic range and boundary conditions for the subsequent characterization of probability distribution. Furthermore, when constructing the probability distribution fuzzy set, the Copula theory can well capture the dependency structure between variables, and then integrate multiple related factors to accurately describe their joint probability distribution characteristics. At the same time, the Wasserstein distance can measure the differences between different probability distributions, and then define the range of the probability distribution fuzzy set, so that the constructed fuzzy set can cover a variety of possible probability distribution situations. Finally, by calculating the initial day-ahead revenue data through the transaction joint scenario set, the theoretically obtainable revenue of the station group under each scenario can be quantified according to different typical transaction situations. Further, on the basis of the initial day-ahead revenue data, the target day-ahead revenue data is calculated by combining the transaction joint scenario set and the probability distribution fuzzy set, taking into account the uncertainty caused by the power error of wind and solar power generation forecast and a variety of possible probability distribution situations, and realizing a more accurate assessment of the revenue of the wind, solar and storage station group. Finally, by calculating the risk value of the revenue condition, the risks faced by the wind, solar and storage station group can be quantified, providing a key scientific basis for operators in making decisions, so that they can more reasonably plan the operation strategy and market trading strategy of the station group according to their risk tolerance and revenue goals, and effectively balance the relationship between revenue acquisition and risk control.

[0113] In this embodiment, a method for optimizing the day-ahead operation plan of a wind-solar-storage station group is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers, etc. Figure 3 is a flow chart of a method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0114] Step S301, obtain the transaction data set of the wind, solar and storage station group participating in the day-ahead energy market and the historical prediction power error data set of the wind, solar and storage station group. Figure 2Step S201 of the illustrated embodiment will not be described in detail here.

[0115] Step S302: Based on the transaction data set and the historical prediction power error data set, establish a transaction joint scenario set and a probability distribution fuzzy set. Figure 2 Step S202 of the illustrated embodiment will not be described in detail here.

[0116] Step S303: Calculate the target day-ahead revenue data and revenue conditional value at risk of the wind, solar and storage station cluster based on the transaction joint scenario set and probability distribution fuzzy set. Figure 2 Step S203 of the illustrated embodiment will not be described in detail here.

[0117] Step S304, based on the transaction joint scenario set, the historical prediction power error data set, the target day-ahead revenue data and the revenue condition risk value, establish an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage station cluster.

[0118] Specifically, the above step S304 includes:

[0119] Step S3041, determining the optimization target based on the historical predicted power error data set, the target day-ahead return data and the return conditional value at risk.

[0120] Specifically, the optimization objective is to maximize the linear combination of the revenue and conditional risk value of the wind-solar-storage station group under the joint probability distribution of the worst wind and solar day-ahead short-term forecast active power errors, as shown in the following equation (13):

[0121]

[0122] Where: γ represents the risk weight factor.

[0123] Step S3042, based on the preset constraint condition set, transaction joint scenario set and optimization target, establish the initial day-ahead distributed blue-robust optimization scheduling model of the wind, solar and storage station cluster.

[0124] The preset constraint condition set may include:

[0125] (1) The wind power and photovoltaic operation constraints of each renewable energy site are shown in the following equations (14) and (15), respectively:

[0126]

[0127] (2) The charging and discharging operation constraints of each energy storage system are shown in the following equations (16) to (23):

[0128]

[0129] Where: They represent the charging / discharging state of the energy storage system of the kth station in the sth scenario at time t; They represent the maximum charging / discharging power of the energy storage system at the kth station in the sth scenario at time t; represents the planned power generation of the energy storage system of the kth station in the sth scenario at time t; They represent the maximum / minimum energy storage capacity of the energy storage system of the kth station at time t respectively; represents the energy storage capacity of the energy storage system of the kth station in the sth scenario to execute the day-ahead operation plan up to time t;

[0130] (3) The day-ahead planned operation constraints of the renewable energy station cluster are shown in the following equations (24) to (26):

[0131]

[0132] Specifically, on the basis of considering the typical joint scenario of day-ahead cleared electricity volume and cleared electricity price, the goal is to maximize the linear combination of the revenue and conditional risk value of the revenue of the wind, solar and storage station cluster under the joint probability distribution of the worst day-ahead short-term predicted active power errors of wind and solar power, and under the constraints of the above-mentioned preset constraint set, an initial day-ahead distributed robust optimization scheduling model for the wind, solar and storage station cluster is constructed.

[0133] Step S305, solving the day-ahead distributed blue-rod optimization scheduling model of the initial wind, solar and storage farm cluster to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm cluster.

[0134] Specifically, the above step S305 includes:

[0135] Step S3051, using linear decision rules and duality theory, reconstruct the initial day-ahead blue-rod optimization scheduling model of the wind, solar and storage farms to obtain the day-ahead blue-rod optimization scheduling model of the target wind, solar and storage farms.

[0136] Among them, the Linear Decision Rule represents a decision-making method that assumes that there is a linear relationship between the decision variables and the objective function and the constraints; Duality Theory is an important theory in linear programming. For every linear programming problem (called the original problem), there is a corresponding dual problem. The original problem and the dual problem are related to each other in mathematical form, and there is a specific relationship between their optimal solutions.

[0137] Specifically, the linear decision rules and duality theory can be used to reconstruct the initial day-ahead blue-robust optimization scheduling model of the wind, solar and storage farm cluster into a form that can be solved by the solver, that is, to reconstruct it into the day-ahead blue-robust optimization scheduling model of the target wind, solar and storage farm cluster.

[0138] (1) Reconstruction of linear decision rules

[0139] Based on the linear decision rule, the decision variables of wind, solar and storage are expressed as the following equations (27) to (30):

[0140]

[0141]

[0142] Where: are the linear decision coefficients of wind, solar, and storage at the kth site in the sth scene at time t, x = 0, 1, ..., 2N WSS .

[0143] Furthermore, the above equations (27) to (30) are substituted into the above preset constraint condition set, and the above preset constraint condition set is represented by the following equation (31):

[0144]

[0145] Furthermore, due to and uncertain variables Has a linear function relationship, Must be at the upper limit and the lower limit δ obtain the optimal value, so the above equation (31) can be reconstructed into the following equations (32) and (33):

[0146]

[0147] H( δ )-G≤0 (33)

[0148] Further, substituting the above equations (27) to (30) into the income And represents a concise expression of the return, where K is the coefficient matrix of the uncertain variables and cst is the constant term. At the same time, a concise expression of the conditional risk value can also be obtained, as shown in the following relationships (34) to (36):

[0149]

[0150] Furthermore, the optimization objective shown in the above equation (13) can be simplified to the following equation (37):

[0151]

[0152] (2) Dual Reconstruction

[0153] First, considering the Wasserstein distance constraint, the inner layer maximization problem can be expressed as the following equation (38):

[0154]

[0155]

[0156] Furthermore, since the empirical distribution is discrete, we can obtain the following relationship (42):

[0157]

[0158] Therefore, according to the above equation (42), the inner layer maximization problem shown in the above equation (38) can be rewritten as shown in the following equation (43):

[0159]

[0160] Furthermore, based on the Lagrange duality theory and the above relation (35), the following relation (46) can be obtained:

[0161]

[0162] Furthermore, combining the following equations (49) and (50):

[0163]

[0164] The initial day-ahead distributed blue-robust optimization scheduling model of the wind-solar-storage station group is reconstructed into a mixed integer second-order cone programming, as shown in the following equation (51):

[0165]

[0166] Step S3052, solving the day-ahead distributed blue-rod optimization scheduling model of the target wind, solar and storage farm group to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm group.

[0167] Specifically, the solver can be used to solve the reconstructed day-ahead distributed blue-rod optimization scheduling model of the target wind, solar and storage farm cluster and output the day-ahead operation plan optimization result of the wind, solar and storage farm cluster.

[0168] The method for optimizing the day-ahead operation plan of a wind, solar, and storage farm cluster provided in this embodiment determines the optimization target by combining the target day-ahead revenue data and the risk value of revenue conditions, breaking the traditional single-dimensional decision-making model that only focuses on maximizing revenue, comprehensively weighing the revenue acquisition and the risk conditions faced during the operation of the wind, solar, and storage farm cluster, and fully recognizing the impact of the volatility that the uncertainty of wind and solar power generation may bring to revenue, so that the optimization target is more in line with actual operation needs. Furthermore, an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar, and storage farm cluster is established based on a set of preset constraints, a set of transaction joint scenarios, and a determined optimization target. By integrating realistic operational constraints such as the wind, solar, and storage operation constraints of the wind, solar, and storage farm cluster and the day-ahead operation plan constraints of the entire station into the model, it is ensured that the operation plan generated by the constructed model is feasible in actual operation, while fully optimizing resource allocation and scheduling strategies under the established goals, thereby improving overall operational efficiency and stability. The model is reconstructed using linear decision rules and duality theory to improve the model structure, making the reconstructed target wind-solar-storage station group's day-ahead distributed blue-rod optimization scheduling model easier to solve and better reflecting the relationship between variables and the balance between optimization objectives and constraints, thereby improving solution efficiency and accuracy.

[0169] In one example, if Figure 4 As shown in the figure, a large-scale wind-solar-storage station group day-ahead operation plan optimization method considering deviation assessment penalty and distributed robust condition risk value is provided, including:

[0170] Step 1: Obtain historical data on the cleared electricity volume, cleared electricity price, and the short-term forecasted active power error of wind and solar power stations participating in the day-ahead energy market.

[0171] Step 2: Considering the multiple uncertainties of the day-ahead cleared electricity quantity, day-ahead cleared electricity price, and the day-ahead short-term forecasted active power errors of wind and solar power, the typical joint scenario of the day-ahead cleared electricity quantity and cleared electricity price is obtained based on the backward reduction method, and the support set of the day-ahead short-term forecasted active power errors of wind and solar power driven by historical data is established. Based on the Copula theory and Wasserstein distance, the probability distribution fuzzy set of the day-ahead short-term forecasted active power errors of wind and solar power considering the spatiotemporal correlation of wind and solar power resources is constructed:

[0172] (1) A typical joint scenario for generating day-ahead clearing electricity quantity and clearing electricity price based on the backward curtailment method;

[0173] (2) Establish a support set of short-term forecast active power errors of wind and solar power based on historical data;

[0174] (3) Considering the spatiotemporal correlation of wind and solar resources at each wind and solar storage station, the probability distribution fuzzy set of the short-term forecast active power error of wind and solar power is constructed based on the Copula theory and Wasserstein distance.

[0175] Step 3: Based on the typical joint scenario of day-ahead cleared electricity quantity and cleared electricity price and the probability distribution fuzzy set of the day-ahead short-term forecast active power error of wind and solar, calculate the day-ahead revenue of the wind, solar and storage station cluster and the conditional risk value of the revenue.

[0176] Step 4: Considering the typical joint scenario of day-ahead cleared electricity volume and cleared electricity price, the goal is to maximize the linear combination of the revenue and conditional risk value of the revenue of the wind-solar-storage farm cluster under the joint probability distribution of the worst day-ahead short-term forecast active power error of wind and solar power, and based on the operation constraints of wind, solar and storage and over-generation constraints, a day-ahead distributed blue optimization scheduling model for the wind-solar-storage farm cluster is constructed.

[0177] Step 5: Based on linear decision rules and duality theory, the day-ahead distributed robust optimization scheduling model is reconstructed into a form that can be solved by the solver.

[0178] The large-scale wind, solar and storage station group day-ahead operation plan optimization method provided in this example considering deviation assessment penalty and distributed blue-bar condition risk value has the following beneficial effects:

[0179] 1. The uncertainty of the day-ahead energy market clearing electricity volume, clearing electricity price and short-term forecast error of renewable energy is considered. The scenario method is used to deal with the uncertainty of the day-ahead clearing electricity volume and the day-ahead clearing electricity price. The distributed robust optimization is used to deal with the uncertainty of renewable energy forecast error. In addition, the spatiotemporal correlation of wind and solar resources at each station within the wind, solar and storage station cluster is considered when generating the probability distribution fuzzy set.

[0180] 2. The total output of the wind and solar power storage station group is allowed to exceed the cleared electricity by a certain proportion, that is, the over-generation margin is taken into account, while the deviation penalty caused by the part exceeding the cleared electricity is taken into account.

[0181] 3. The risk of income is taken into consideration, and the distributed robust conditional value at risk is used to measure the risk. The dispatchers can choose the appropriate risk weight factor according to their own risk preference to balance the income and risk of the day-ahead dispatch optimization of the wind, solar and storage station cluster.

[0182] Therefore, compared with the traditional day-ahead operation plan optimization method, the scheme designed in this example can be applied to the day-ahead operation plan optimization of a station group composed of multiple renewable energy power stations. In addition, this scheme can ensure the economic and robustness of the day-ahead operation plan of the wind, solar and storage station group, and improve the utilization rate of the energy storage system of each station and the ability to respond to electricity price arbitrage.

[0183] In this embodiment, a device for optimizing the day-ahead operation plan of a wind-solar-storage station group is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0184] This embodiment provides a device for optimizing the day-ahead operation plan of a wind-solar-storage station group. Figure 5 As shown, the device comprises:

[0185] The acquisition module 501 is used to acquire the transaction data set of the wind, solar and storage farm cluster participating in the day-ahead energy market and the historical prediction power error data set of the wind, solar and storage farm cluster.

[0186] The first establishment module 502 is used to establish a transaction joint scenario set and a probability distribution fuzzy set based on the transaction data set and the historical prediction power error data set. The probability distribution fuzzy set is used to characterize the probability distribution of the short-term predicted active power error of wind and solar power in the day before considering the spatiotemporal correlation of wind and solar power resources.

[0187] The calculation module 503 is used to calculate the target day-ahead revenue data and the revenue conditional risk value of the wind, solar and storage station group based on the transaction joint scenario set and the probability distribution fuzzy set.

[0188] The second establishment module 504 is used to establish an initial day-ahead distributed blue-rod optimization scheduling model for the wind, solar and storage station cluster based on the transaction joint scenario set, the historical prediction power error data set, the target day-ahead revenue data and the revenue condition risk value.

[0189] The solution module 505 is used to solve the day-ahead distributed blue-rod optimization scheduling model of the initial wind, solar and storage farm cluster to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm cluster.

[0190] In some optional implementations, the acquisition module 501 includes:

[0191] The acquisition submodule is used to obtain the historical cleared electricity data and historical cleared electricity price data of the wind-solar-storage station group participating in the day-ahead energy market, the historical data of the wind-solar-storage station group's short-term predicted active power of the wind and solar power in the day-ahead, and the real-time historical data of the maximum active power that can be generated by the wind and solar power.

[0192] The first determination submodule is used to determine the transaction data set based on the historical clearing electricity quantity data and the historical clearing electricity price data.

[0193] The second determination submodule is used to determine a historical prediction power error data set based on the historical data and the real-time historical data.

[0194] In some optional implementations, the first establishing module 502 includes:

[0195] The generation submodule is used to generate a set of transaction joint scenarios based on the transaction data set after processing by the backward reduction method.

[0196] The first establishing submodule is used to establish a predicted power error support set based on a historical predicted power error data set.

[0197] The second submodule is used to establish a probability distribution fuzzy set based on the predicted power error support set using Copula theory and Wasserstein distance.

[0198] In some optional implementations, the calculation module 503 includes:

[0199] The first calculation submodule is used to calculate the initial day-ahead revenue data of the wind, solar and storage station cluster based on the transaction joint scenario set.

[0200] The second calculation submodule is used to calculate the target day-ahead revenue data of the wind, solar and storage station cluster based on the initial day-ahead revenue data, the transaction joint scenario set and the probability distribution fuzzy set.

[0201] The third calculation submodule is used to calculate the return conditional risk value based on the target day return data.

[0202] In some optional implementations, the second establishing module 504 includes:

[0203] The third determination submodule is used to determine the optimization target based on the historical prediction power error data set, the target day-ahead return data and the return conditional risk value.

[0204] The third submodule is used to establish an initial day-ahead distributed optimization scheduling model for the wind, solar and storage station cluster based on a preset set of constraints, a set of transaction joint scenarios and optimization objectives.

[0205] In some optional implementations, the solution module 505 includes:

[0206] The reconstruction submodule is used to reconstruct the initial day-ahead blue-rod optimization scheduling model of the wind-solar-storage station group by using linear decision rules and duality theory, and obtain the day-ahead blue-rod optimization scheduling model of the target wind-solar-storage station group.

[0207] The solution submodule is used to solve the day-ahead distributed blue-rod optimization scheduling model of the target wind, solar and storage farm group, and obtain the day-ahead operation plan optimization result of the wind, solar and storage farm group.

[0208] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0209] The wind, solar and storage station group day-ahead operation plan optimization device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0210] The embodiment of the present invention also provides a computer device having the above Figure 5 The device for optimizing the day-ahead operation plan of a group of wind, solar and storage stations is shown.

[0211] See also Figure 6 , Figure 6 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 6 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0212] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0213] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0214] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0215] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0216] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0217] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0218] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0219] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for optimizing the day-ahead operation plan of a wind-solar-storage station group, characterized in that: The method comprises: Obtain the transaction data set of the wind, solar and storage station cluster participating in the day-ahead energy market and the historical forecast power error data set of the wind, solar and storage station cluster; Based on the transaction data set and the historical predicted power error data set, a transaction joint scenario set and a probability distribution fuzzy set are established, and the probability distribution fuzzy set is used to characterize the probability distribution of the day-ahead short-term predicted active power errors of wind and solar power taking into account the spatiotemporal correlation of wind and solar power resources; Based on the transaction joint scenario set and the probability distribution fuzzy set, the target day-ahead revenue data and the revenue conditional risk value of the wind, solar and storage station group are calculated; Based on the transaction joint scenario set, the historical predicted power error data set, the target day-ahead revenue data and the revenue condition risk value, an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage station cluster is established; The day-ahead distributed blue-rod optimization scheduling model of the initial wind-solar-storage farm cluster is solved to obtain the day-ahead operation plan optimization result of the wind-solar-storage farm cluster.

2. The method according to claim 1, characterized in that: Obtain the transaction data set and historical power forecast error data set of wind, solar and storage station clusters participating in the day-ahead energy market, including: Obtain the historical cleared electricity data and historical cleared electricity price data of the wind-solar-storage station group participating in the day-ahead energy market, the historical data of the short-term predicted active power of the wind and solar-solar day-ahead of the wind-solar-storage station group, and the real-time historical data of the maximum active power that can be generated by the wind and solar-solar; Determining the transaction data set based on the historical clearing electricity quantity data and the historical clearing electricity price data; The historical predicted power error data set is determined based on the historical data and the real-time historical data.

3. The method according to claim 1, characterized in that Based on the transaction data set and the historical prediction power error data set, a transaction joint scenario set and a probability distribution fuzzy set are established, including: Based on the transaction data set, the transaction joint scenario set is generated by processing through a backward reduction method; Based on the historical predicted power error data set, establishing a predicted power error support set; Based on the predicted power error support set, the probability distribution fuzzy set is established by using Copula theory and Wasserstein distance.

4. The method according to claim 1, characterized in that Based on the transaction joint scenario set and the probability distribution fuzzy set, the target day-ahead revenue data and the revenue conditional risk value of the wind, solar and storage station group are calculated, including: Based on the transaction joint scenario set, calculating the initial day-ahead revenue data of the wind, solar and storage station group; Calculate the target day-ahead revenue data of the wind-solar-storage station group based on the initial day-ahead revenue data, the transaction joint scenario set and the probability distribution fuzzy set; The return conditional value at risk is calculated based on the target day-ahead return data.

5. The method according to claim 1, characterized in that Based on the transaction joint scenario set, the historical prediction power error data set, the target day-ahead revenue data and the revenue condition risk value, an initial day-ahead distributed blue-robust optimization scheduling model for the wind, solar and storage station cluster is established, including: Determining an optimization target based on the historical predicted power error data set, the target day-ahead return data, and the return conditional value at risk; Based on the preset constraint condition set, the transaction joint scenario set and the optimization target, a day-ahead distributed blue-robust optimization scheduling model for the initial wind-solar-storage station cluster is established.

6. The method according to claim 1, characterized in that Solve the day-ahead distributed blue-rod optimization dispatch model of the initial wind-solar-storage farm group to obtain the day-ahead operation plan optimization result of the wind-solar-storage farm group, including: By using linear decision rules and duality theory, the initial wind-solar-storage farm group day-ahead distributed blue-rod optimization dispatching model is reconstructed to obtain the target wind-solar-storage farm group day-ahead distributed blue-rod optimization dispatching model; The day-ahead distributed blue-rod optimization scheduling model of the target wind, solar and storage farm station group is solved to obtain the day-ahead operation plan optimization result of the wind, solar and storage farm station group.

7. A device for optimizing the day-ahead operation plan of a wind-solar-storage station group, characterized in that: The device comprises: An acquisition module is used to acquire a transaction data set of the wind, solar and storage farm cluster participating in the day-ahead energy market and a historical prediction power error data set of the wind, solar and storage farm cluster; A first establishment module is used to establish a transaction joint scenario set and a probability distribution fuzzy set based on the transaction data set and the historical prediction power error data set, wherein the probability distribution fuzzy set is used to characterize the probability distribution of the short-term prediction active power error of wind and solar energy in the day ahead considering the spatiotemporal correlation of wind and solar energy resources; A calculation module, used for calculating the target day-ahead revenue data and the revenue conditional risk value of the wind-solar-storage station group based on the transaction joint scenario set and the probability distribution fuzzy set; The second establishment module is used to establish an initial wind-solar-storage station group day-ahead distributed blue-robust optimization scheduling model based on the transaction joint scenario set, the historical predicted power error data set, the target day-ahead revenue data and the revenue condition risk value; The solution module is used to solve the day-ahead distributed blue-rod optimization scheduling model of the initial wind-solar-storage station group to obtain the day-ahead operation plan optimization result of the wind-solar-storage station group.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to any one of claims 1 to 6.

10. A computer program product, characterized in that It includes computer instructions, and the computer instructions are used to enable a computer to execute the method for optimizing the day-ahead operation plan of a wind-solar-storage station group according to any one of claims 1 to 6.

Images