Wind power asset transaction management system based on centralized large storage

Abstract

The invention discloses a wind power asset transaction management system based on centralized large storage, relates to the technical field of electric power asset transaction, and solves the problems that a transaction strategy is fragile when facing policy adjustment, and the transaction effect is poor due to the fact that a traditional electric power transaction system cannot fully integrate evaluation of the influence of policy changes on the market. And the capability of monitoring in real time and dynamically adjusting the transaction strategy is lacked. The system comprises a data acquisition and analysis unit, a risk income analysis unit, an auxiliary decision-making unit and a terminal, can acquire new energy power generation data, historical transaction data and market disclosure data and perform further data operation, and takes maximization of the income after risk adjustment as a target, considers the influence of policy change, constructs a target function, and performs decision-making on the new energy power generation data, the historical transaction data and the market disclosure data. Therefore, the decision can reduce the influence caused by the policy change, and maximizes the benefit after the risk adjustment. The method can comprehensively consider multiple factors such as market supply and demand, price fluctuation, policy change and the like, and provides scientific transaction decision support for wind power enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of power asset trading, and in particular to a wind power asset trading management system based on centralized large storage. Background Art

[0003] Traditional wind power asset trading relies on manual decision-making and limited data, lacking systematic and scientific management methods. In dealing with the complexity of market dynamics and policy changes, some traditional power trading systems fail to fully integrate the assessment of the impact of policy changes on the market, making trading strategies vulnerable in the face of policy adjustments, and thus lacking the ability to monitor and dynamically adjust trading strategies in real time, which is a major flaw in the rapidly changing power market.

[0004] In view of this, a wind power asset transaction management system based on centralized large storage is needed. Summary of the invention

[0005] In view of the problem that the traditional power trading system in the prior art fails to fully integrate the assessment of the impact of policy changes on the market, making the trading strategy vulnerable in the face of policy adjustments, and thus lacking the ability to monitor in real time and dynamically adjust the trading strategy, the present invention provides a wind power asset trading management system based on centralized large storage, which can collect new energy power generation data, historical trading data and market disclosure data and perform further data calculations, so that these data indicators can be quantified. In addition, it also constructs an objective function with the goal of maximizing risk-adjusted returns, and considers the impact of policy changes in the process, so that decisions can further reduce the impact of policy changes and maximize risk-adjusted returns. This solution can comprehensively consider multiple factors such as market supply and demand, price fluctuations, and policy changes, and provide scientific trading decision support for wind power companies. The specific technical solution is as follows:

[0006] A wind power asset transaction management system based on centralized large storage, comprising:

[0007] The data collection and analysis unit includes a data collection module, a data analysis module and a data prediction module. The data collection module collects new energy power generation data, historical transaction data, energy storage cost, load loss cost and market disclosure data. The data analysis module further calculates based on the data collected by the data collection module and transmits the calculation results to the risk-benefit analysis unit for processing; the data prediction module predicts future data based on historical data based on the trained prediction model;

[0008] A risk-return analysis unit, which aims to maximize risk-adjusted returns and constructs an objective function, wherein the influencing factors of the objective function include returns, prices, and policy costs;

[0009] A decision-making support unit, used to solve the objective function constructed by the risk-benefit analysis unit;

[0010] The terminal is electrically connected to the auxiliary decision unit and is used to receive output data of the auxiliary decision unit.

[0011] Preferably, the new energy power generation data includes power generation; the historical transaction data includes transaction price and transaction volume; the market disclosure data includes the number of policy changes and the cost before the policy change and the cost after the policy change.

[0012] Preferably, the data analysis unit performs the following calculations:

[0013] Historical average transaction price:

[0014]

[0015] Among them, P i is the price of the ith transaction, and N is the number of transactions;

[0016] Standard deviation of price:

[0017]

[0018] Among them, P i is the price of the ith transaction, P avg is the historical average transaction price;

[0019] Policy change rate:

[0020]

[0021] Among them, R c is the number of changes, T is the total time period;

[0022] Policy cost impact coefficient:

[0023]

[0024] Among them, C before is the cost before the policy change, C after It is the cost after the policy change.

[0025] Preferably, the objective function is as follows:

[0026] Maximize Z=Π-λσ price -μC policy -ηC 失负荷

[0027] Among them, Π is the expected return, λ is the risk aversion coefficient, σ price is the standard deviation of price, μ is the policy cost sensitivity coefficient, C policyis the policy cost impact coefficient, η is the load loss impact coefficient, C 失负荷 is the loss of load cost;

[0028] Π=P i ×Q(P i )-C total ;

[0029] Among them, P i is the electricity price, Q(P i ) is the same as P i The corresponding amount of electricity purchased, C total is the total cost;

[0030]

[0031] Among them, P i is the price of the ith transaction, P avg The historical average transaction price

[0032]

[0033] Among them, C before is the cost before the policy change, C after It is the cost after the policy change.

[0034] Preferably, the total cost includes wind power cost and energy storage cost, as shown below:

[0035] C total =C 风电 +C 储能 ;

[0036] In the formula, C 风电 is the cost of wind power, C 储能 The cost of energy storage;

[0037] The wind power cost is calculated by the following formula:

[0038]

[0039] Among them, P dynamic_cost is the dynamic investment cost of wind farm, T O&M is the operation time of the wind farm, D deprectation is the depreciation of fixed assets of the wind farm, P O&M is the operation and maintenance cost of the wind farm, R tax is the income tax rate for wind farms, R discount is the discount rate, V restdualvaiue is the residual value of fixed assets of the wind farm, E annual is the annual power generation of the wind farm.

[0040] The energy storage cost is calculated by the following formula:

[0041]

[0042] Where λ is the efficiency of the energy storage system, P c is the rated power of the energy storage system, T is the life cycle of the project, n c (t) is the number of cycles in year t, r is the discount rate, O&M(t) is the operation and maintenance cost in year t, C E is the energy storage cost of the energy storage system, C P is the power storage cost of the energy storage system, and d is the service life of the energy storage system.

[0043] Preferably, the following constraints are set:

[0044] Budget Constraint:

[0045] C total ≤B;

[0046] Among them, C total is the total cost, including generation cost, storage cost and transaction cost; B is the available budget;

[0047] Generation capacity constraints:

[0048] G renewable ≤G eap ;

[0049] Among them, G renewable is the total power generation in the cycle, G eap is the upper limit of power generation capacity;

[0050] Market volume constraints:

[0051] V min ≤V≤V max ;

[0052] Among them, V is the market trading volume, V min and V max They are the minimum and maximum limits on market trading volume respectively.

[0053] Preferably, the auxiliary decision unit uses a gradient ascent method to solve the objective function, and the specific steps are as follows:

[0054] Step 1: Obtain the electricity price-purchase quantity dataset at future times and future energy storage capacity-load loss cost dataset Randomly select an initial point As a starting point;

[0055] Step 2: At the current point Calculate the gradient of the objective function Z, that is:

[0056]

[0057] Step 3: Update the parameters along the direction of the gradient, that is, Where α is the learning rate, a positive decimal that controls the step size;

[0058] Step 4: Check whether the magnitude of the gradient is less than a preset threshold, or check whether the value of the objective function changes very little in several consecutive iterations. If so, the algorithm is considered to have converged.

[0059] Step 5: If there is no convergence, return to step 2 and continue iterating;

[0060] Step 6: When the algorithm converges, output the current The value of .

[0061] Preferably, the training and data processing process of the prediction model is as follows:

[0062] Define the electricity price sequence P and energy storage capacity sequence E, which can be expressed as: P = {P min ,P1,P2,···,P max}; E = {E min ,E1,E2,···,E max}, where the electricity price sequence P and the energy storage capacity sequence E are both one-time increasing sequences, which can be set as an arithmetic sequence or a non-arithmetic sequence. When set as an arithmetic sequence, the interval of the arithmetic sequence can be set. The narrower the interval, the higher the accuracy.

[0063] Get the historical electricity price P t , historical electricity purchase amount Q t , historical energy storage capacity E t and the historical load loss cost C t ;

[0064] The combination of historical electricity price and energy storage capacity is taken as feature X, that is, P t and E t Pair to form a feature vector; the historical electricity purchase volume Q t As the purchased electricity quantity label corresponding to the electricity price, the historical load loss cost C t As a load cost label corresponding to the energy storage capacity;

[0065] Divide the features and labels into training and test sets, for example, 80% training set and 20% test set. Standardize the features of the training and test sets to improve the efficiency and performance of model training. Standardize the labels of the training set, and after prediction, de-normalize the prediction results to obtain the actual values.

[0066] Select a regression model (Random Forest Regressor) and use the features and standardized labels on the training set to train two models: one for predicting the amount of electricity purchased corresponding to the electricity price, and the other for predicting the load cost corresponding to the energy storage capacity. Use the features on the test set to make predictions. Calculate the error between the prediction result and the actual label, such as the mean square error (MSE)

[0067] For each electricity price in the electricity price sequence P and each energy storage capacity in the energy storage capacity sequence E, the trained model is used to predict the corresponding power purchase and load cost.

[0068] For each electricity price in the electricity price sequence P and each energy storage capacity in the energy storage capacity sequence E, the trained model is used to predict the corresponding amount of electricity purchased and load cost. That is, the trained model outputs the amount of electricity purchased corresponding to each electricity price in the future electricity price sequence. And the load cost corresponding to each energy storage capacity in the energy storage capacity sequence at the future moment

[0069] Combine the prediction results into two data sets: future electricity price-purchased electricity data set and future energy storage capacity-load loss cost dataset

[0070] A computer-readable storage medium includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to execute the wind power asset trading management system based on centralized large storage as described above.

[0071] A processor is used to run a program, wherein when the program is run, the wind power asset trading management system based on centralized large storage as described above is executed.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] 1. The present invention provides a comprehensive data view for decision makers by acquiring new energy power generation data in real time, combining historical transaction data and market disclosure data. This data integration not only includes basic information on wind power output and market transactions, but also involves changes in market rules and policies, which are key factors affecting wind power asset transactions. By quantifying these data, the solution can calculate key indicators such as the historical average transaction price, transaction volume, and standard deviation of prices, providing a solid data foundation for trading strategies. In addition, the introduction of policy change rate and policy cost impact coefficient in the solution enables the system to predict and adapt to the impact of policy changes on the market, thereby making more flexible and adaptable trading decisions in a changing market environment.

[0074] 2. The present invention takes into account expected returns, price volatility risks, and policy cost impacts by constructing an objective function, so that the trading strategy is no longer a single pursuit of maximizing returns, but a balance between returns and risks. This balance is achieved by introducing a risk aversion coefficient and a policy cost sensitivity coefficient. These two parameters allow decision makers to adjust strategies based on their own risk preferences and sensitivity to policy changes. For example, if market volatility increases, the risk aversion coefficient can be increased accordingly to reduce risk exposure; similarly, if policy changes have a greater impact on costs, adjustments to the policy cost sensitivity coefficient will help companies mitigate the adverse effects of policy changes. This dynamic adjustment mechanism makes wind power asset transactions more robust and can find the optimal transaction path in uncertainty.

[0075] 3. The present invention enables enterprises to quickly seize market opportunities by monitoring market and policy changes in real time and dynamically adjusting trading strategies. This rapid response capability is due to the data analysis and quantitative models integrated in the solution, which can process and analyze large amounts of data in a short period of time and provide immediate decision support. In addition, the automated nature of the solution reduces the need for manual operations, reduces the possibility of human errors, and improves the accuracy and efficiency of transactions. In the highly competitive electricity market, this fast and accurate trading capability is the key to companies gaining an advantage. Through this technical solution, wind power companies can not only better manage their assets, but also occupy a more advantageous position in the market and improve their overall market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0077] Figure 1 It is a system principle diagram of the present invention. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0080] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0081] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0082] In one embodiment of the present invention, a wind power asset transaction management system based on centralized large storage is provided, such as Figure 1 As shown, including:

[0083] 1. Data collection and analysis unit, including data collection module, data analysis module and data prediction module.

[0084] The data collection module collects new energy power generation data, historical transaction data, energy storage costs, load loss costs, and market disclosure data. Among them, new energy power generation data is obtained through self-owned IoT devices or power grid company data centers, which are real-time wind power and photovoltaic power generation data, including power generation. For example, the wind power generation data for the past month was obtained from the power grid company, and the total power generation was 100,000MWh. Historical transaction data includes transaction prices and transaction volumes, which are obtained from the power trading platform. Market disclosure data includes market rules and policies, which are obtained from the official website of the power market regulatory agency.

[0085] The data analysis module receives the raw data collected by the data acquisition module and performs quantitative processing on it, thereby forming data that can be used for numerical calculations for use by the risk-benefit analysis unit. The data analysis module mainly outputs the following data:

[0086] 1. Historical average transaction price:

[0087]

[0088] Among them, P i is the price of the ith transaction, and N is the number of transactions:

[0089] The historical average transaction price reflects the average price level of the market in a certain period of time and is the basis for calculating expected returns. As the basis for calculating expected returns, it directly affects the valuation and trading decisions of wind power assets.

[0090] 2. Historical average trading volume:

[0091]

[0092] Among them, V i is the trading volume on the ith trading day, and N is the number of transactions;

[0093] Historical average trading volume is an important data indicator that reflects market activity, capital inflows and outflows, and investor sentiment. In wind power asset transactions, higher trading volume may mean that the market has a higher interest in wind power assets, which may affect trading strategies and pricing. Historical average trading volume is used to measure the volatility of market prices and is a key indicator for risk management. In the objective function, it affects the calculation of expected returns through risk adjustment items, helping companies evaluate and manage market risks.

[0094] 3. Standard deviation of price:

[0095]

[0096] Among them, P i is the price of the ith transaction, P avg is the historical average transaction price. The standard deviation of price is an indicator of market price volatility, which reflects the degree to which the price deviates from its average value. Subtracting the standard deviation of price from the objective function can help us quantify market risk, so that we can take into account the impact of risk while pursuing returns.

[0097] 4. Policy change rate:

[0098]

[0099] Among them, R c is the number of changes, T is the total time period;

[0100] 5. Policy cost impact coefficient:

[0101]

[0102] Among them, C before is the cost before the policy change, C after It is the cost after the policy change.

[0103] The data prediction module predicts future data based on historical data based on the trained prediction model. The details are as follows:

[0104] First, define the electricity price sequence P and energy storage capacity sequence E, which can be expressed as: P = {P min ,P1,P2,···,P max}; E = {E min ,E1,E2,···,E max}, where the electricity price sequence P and the energy storage capacity sequence E are both one-time increasing sequences, which can be set as an arithmetic sequence or a non-arithmetic sequence. When set as an arithmetic sequence, the interval of the arithmetic sequence can be set. The narrower the interval, the higher the accuracy.

[0105] Get the historical electricity price P t , historical electricity purchase amount Q t , historical energy storage capacity E t and the historical load loss cost C t ;

[0106] The combination of historical electricity price and energy storage capacity is taken as feature X, that is, P t and E t Pair to form a feature vector; the historical electricity purchase volume Q t As the purchased electricity quantity label corresponding to the electricity price, the historical load loss cost C t As a load cost label corresponding to the energy storage capacity;

[0107] Divide the features and labels into training and test sets, for example, 80% training set and 20% test set. Standardize the features of the training and test sets to improve the efficiency and performance of model training. Standardize the labels of the training set, and after prediction, de-normalize the prediction results to obtain the actual values.

[0108] Select a regression model (Random Forest Regressor) and use the features and standardized labels on the training set to train two models: one for predicting the amount of electricity purchased corresponding to the electricity price, and the other for predicting the load cost corresponding to the energy storage capacity. Use the features on the test set to make predictions. Calculate the error between the prediction result and the actual label, such as the mean square error (MSE)

[0109] For each electricity price in the electricity price sequence P and each energy storage capacity in the energy storage capacity sequence E, the trained model is used to predict the corresponding power purchase and load cost.

[0110] For each electricity price in the electricity price sequence P and each energy storage capacity in the energy storage capacity sequence E, the trained model is used to predict the corresponding amount of electricity purchased and load cost. That is, the trained model outputs the amount of electricity purchased corresponding to each electricity price in the future electricity price sequence. And the load cost corresponding to each energy storage capacity in the energy storage capacity sequence at the future moment

[0111] Combine the prediction results into two data sets: future electricity price-purchased electricity data set and future energy storage capacity-load loss cost dataset

[0112] Second, the risk-return analysis unit aims to maximize risk-adjusted returns, construct an objective function, and consider the impact of policy changes in the process. The objective function is as follows:

[0113] Maximize Z=Π-λσ price -μC policy -ηC 失负荷

[0114] Among them, Π is the expected return, λ is the risk aversion coefficient, σ price is the standard deviation of price, μ is the policy cost sensitivity coefficient, C policy is the policy cost impact coefficient, η is the load loss impact coefficient, C 失负荷 is the loss of load cost. 失负荷 Specifically, it is the probability that the available capacity of the power generation system cannot meet the annual maximum load demand of the system.

[0115] Furthermore, the expected return can be derived as follows:

[0116] Π=P i ×Q(P i )-C total ;

[0117] Among them, P i is the electricity price, Q(P i ) is the same as P i The corresponding amount of electricity purchased, C total is the total cost.

[0118] The total cost includes wind power cost and energy storage cost, as shown below:

[0119] C total =C 风电 +C 储能 ;

[0120] In the formula, C 风电 is the cost of wind power, C 储能 The cost of energy storage;

[0121] The wind power cost is calculated by the following formula:

[0122]

[0123] Among them, P dynamic_cost is the dynamic investment cost of wind farm, T O&M is the operation time of the wind farm, Ddeprectation is the depreciation of fixed assets of the wind farm, P O&M is the operation and maintenance cost of the wind farm, R tax is the income tax rate for wind farms, R discount is the discount rate, is the residual value of fixed assets of the wind farm, E annual is the annual power generation of the wind farm.

[0124] The energy storage cost is calculated by the following formula:

[0125]

[0126] Where λ is the efficiency of the energy storage system, P c is the rated power of the energy storage system, T is the life cycle of the project, n c (t) is the number of cycles in year t, r is the discount rate, O&M(t) is the operation and maintenance cost in year t, C E is the energy storage cost of the energy storage system, C P is the power storage cost of the energy storage system, and d is the service life of the energy storage system.

[0127] Based on the objective function, the following constraints are set:

[0128] 1. Budget Constraints:

[0129] C total ≤B;

[0130] Among them, C total is the total cost, including power generation cost, energy storage cost and transaction cost; B is the available budget.

[0131] 2. Power generation capacity constraints:

[0132] G renewable ≤G eap ;

[0133] Among them, G renewable is the total power generation in the cycle, G eap It is the upper limit of power generation capacity.

[0134] 3. Market transaction volume constraints:

[0135] V min ≤V≤V max ;

[0136] Among them, V is the market trading volume, V min and V max They are the minimum and maximum limits on market trading volume respectively.

[0137] These constraints ensure that the system does not exceed budget, capacity, or policy limits while maximizing risk-adjusted returns. By incorporating these constraints into the objective function, we can find the optimal solution that meets both business goals and actual operating conditions.

[0138] 3. Auxiliary decision-making unit, solving the objective function.

[0139] The auxiliary decision unit is preset with a gradient ascent algorithm. The gradient ascent method is an iterative method that gradually adjusts parameters along the direction of the objective function gradient to find the maximum value of the function. The following are the specific steps of the gradient ascent method:

[0140] Step 1: Obtain the electricity price-purchase quantity dataset at future times and future energy storage capacity-load loss cost dataset Randomly select an initial point As a starting point;

[0141] Step 2: At the current point Calculate the gradient of the objective function Z, that is:

[0142]

[0143] Step 3: Update the parameters along the direction of the gradient, that is, Where α is the learning rate, a positive decimal that controls the step size;

[0144] Step 4: Check whether the magnitude of the gradient is less than a preset threshold, or check whether the value of the objective function changes very little in several consecutive iterations. If so, the algorithm is considered to have converged.

[0145] Step 5: If there is no convergence, return to step 2 and continue iterating;

[0146] Step 6: When the algorithm converges, output the current These values ​​are approximate solutions that maximize the objective function Z, which can determine the strategy for electricity pricing and whether the energy storage capacity changes.

[0147] 4. The terminal is connected to the auxiliary decision-making unit and the data analysis module to receive the output data of the auxiliary decision-making unit and the data analysis module, further affecting the final decision made by the terminal, or providing a reference for the terminal to make the final decision.

[0148] In summary, the present invention first provides a comprehensive data view for decision makers by acquiring new energy power generation data in real time, combining historical transaction data and market disclosure data. This data integration not only includes the basic information of wind power output and market transactions, but also involves changes in market rules and policies, which are key factors affecting wind power asset transactions. By quantifying these data, the scheme can calculate key indicators such as the historical average transaction price, transaction volume, and standard deviation of prices, providing a solid data foundation for trading strategies. In addition, the introduction of the policy change rate and policy cost impact coefficient in the scheme enables the system to predict and adapt to the impact of policy changes on the market, thereby making more flexible and adaptable trading decisions in a changing market environment. Secondly, by constructing an objective function, the present invention takes into account the expected return, price volatility risk, and policy cost impact, so that the trading strategy is no longer a single pursuit of maximizing returns, but a balance between returns and risks. This balance is achieved by introducing the risk aversion coefficient and the policy cost sensitivity coefficient. These two parameters allow decision makers to adjust their strategies according to their own risk preferences and sensitivity to policy changes. For example, if market volatility increases, the risk aversion coefficient can be increased accordingly to reduce risk exposure; similarly, if policy changes have a greater impact on costs, adjustments to the policy cost sensitivity coefficient will help companies mitigate the adverse effects of policy changes. This dynamic adjustment mechanism makes wind power asset transactions more robust and can find the optimal transaction path in uncertainty. In addition, the present invention enables companies to quickly seize market opportunities by monitoring market and policy changes in real time and dynamically adjusting trading strategies. This rapid response capability is due to the data analysis and quantitative models integrated in the solution, which can process and analyze large amounts of data in a short period of time and provide immediate decision support. In addition, the automated nature of the solution reduces the need for manual operations, reduces the possibility of human errors, and improves the accuracy and efficiency of transactions. In the highly competitive electricity market, this fast and accurate trading capability is the key to companies gaining an advantage. Through this technical solution, wind power companies can not only better manage their assets, but also occupy a more advantageous position in the market and improve their overall market competitiveness.

[0149] Those of ordinary skill in the art will appreciate that the units 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 of each example has 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 the present invention.

[0150] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0151] In addition, each functional unit in each embodiment of the present invention 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.

[0152] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A wind power asset trading management system based on centralized large storage, characterized in that: include: A data acquisition and analysis unit, including a data acquisition module, a data analysis module and a data prediction module. The data acquisition module collects new energy power generation data, historical transaction data, energy storage costs, load loss costs and market disclosure data. The data analysis module further calculates based on the data collected by the data acquisition module and transmits the calculation results to the risk-benefit analysis unit for processing; The data prediction module predicts future data based on historical data based on the trained prediction model; A risk-return analysis unit, which aims to maximize risk-adjusted returns and constructs an objective function, wherein the influencing factors of the objective function include returns, prices, and policy costs; A decision-making support unit, used to solve the objective function constructed by the risk-benefit analysis unit; The terminal is electrically connected to the auxiliary decision unit and is used to receive output data of the auxiliary decision unit.

2. A wind power asset transaction management system based on centralized large storage according to claim 1, characterized in that: New energy power generation data includes power generation; historical transaction data includes transaction price and transaction volume; market disclosure data includes the number of policy changes and the cost before and after the policy changes.

3. A centralized wind power asset trading management system based on large storage according to claim 2, characterized in that: The data analysis unit performs the following calculations: Historical average transaction price: Among them, P i is the price of the ith transaction, and N is the number of transactions; Standard deviation of price: Among them, P i is the price of the ith transaction, P avg is the historical average transaction price; Policy change rate: Among them, R c is the number of changes, T is the total time period; Policy cost impact coefficient: Among them, C before is the cost before the policy change, C after It is the cost after the policy change.

4. A wind power asset transaction management system based on centralized large storage according to claim 1, characterized in that: The objective function is as follows: Maximize Z=∏-λσ price -μC policy -ηC 失负荷 Among them, Π is the expected return, λ is the risk aversion coefficient, σ price is the standard deviation of price, μ is the policy cost sensitivity coefficient, C policy is the policy cost impact coefficient, η is the load loss impact coefficient, C 失负荷 is the loss of load cost; P=P i ×Q(P i )-C total ; Among them, P i is the electricity price, Q(P i ) is the same as P i The corresponding amount of electricity purchased, C total is the total cost; Among them, P i is the price of the ith transaction, P avg The historical average transaction price Among them, C before is the cost before the policy change, C after It is the cost after the policy change.

5. A centralized wind power asset trading management system based on large storage according to claim 4, characterized in that: The total cost includes wind power cost and energy storage cost, as shown below: C total =C 风电 +C 储能 ; In the formula, C 风电 is the cost of wind power, C 储能 The cost of energy storage; The wind power cost is calculated by the following formula: Among them, P dynamic_cost is the dynamic investment cost of wind farm, T O&M is the operation time of the wind farm, D deprectation is the depreciation of fixed assets of the wind farm, P O&M is the operation and maintenance cost of the wind farm, R tax is the income tax rate for wind farms, R discount is the discount rate, V restdualvaiue is the residual value of fixed assets of the wind farm, E annual is the annual power generation of the wind farm; The energy storage cost is calculated by the following formula: Where λ is the efficiency of the energy storage system, P c is the rated power of the energy storage system, T is the life cycle of the project, n c (t) is the number of cycles in year t, r is the discount rate, O&M(t) is the operation and maintenance cost in year t, C E is the energy storage cost of the energy storage system, C P is the power storage cost of the energy storage system, and d is the service life of the energy storage system.

6. A centralized wind power asset trading management system based on large storage according to claim 5, characterized in that: Set the following constraints: Budget Constraint: C total ≤B; Among them, C total is the total cost, including generation cost, storage cost and transaction cost; B is the available budget; Generation capacity constraints: G renewable ≤G eap ; Among them, G renewable is the total power generation in the cycle, G eap is the upper limit of power generation capacity; Market volume constraints: In min ≤V≤V max ; Among them, V is the market trading volume, V min and V max They are the minimum and maximum limits on market trading volume respectively.

7. A centralized wind power asset trading management system based on large storage according to claim 4, characterized in that: The auxiliary decision unit uses the gradient ascent method to solve the objective function. The specific steps are as follows: Step 1: Obtain the electricity price-purchase quantity dataset at future times and future energy storage capacity-load loss cost dataset Randomly select an initial point As a starting point; Step 2: At the current point Calculate the gradient of the objective function Z, that is: Step 3: Update the parameters along the direction of the gradient, that is, Where α is the learning rate, a positive decimal that controls the step size; Step 4: Check whether the magnitude of the gradient is less than a preset threshold, or check whether the value of the objective function changes very little in several consecutive iterations. If so, the algorithm is considered to have converged. Step 5: If there is no convergence, return to step 2 and continue iterating; Step 6: When the algorithm converges, output the current The value of .

8. A centralized wind power asset trading management system based on large storage according to claim 1, characterized in that: The training and data processing process of the prediction model is as follows: Define the electricity price sequence P and energy storage capacity sequence E, which can be expressed as: P = {P min ,P1,P2,···,P max }; E = {E min ,E1,E2,···,E max }, where the electricity price sequence P and the energy storage capacity sequence E are both increasing sequences; Get the historical electricity price P t , historical electricity purchase amount Q t , historical energy storage capacity E t and the historical load loss cost C t ; The combination of historical electricity price and energy storage capacity is taken as feature X, that is, P t and E t Pair to form a feature vector; the historical electricity purchase volume Q t As the purchased electricity quantity label corresponding to the electricity price, the historical load loss cost C t As a load cost label corresponding to the energy storage capacity; Divide the features and labels into training sets and test sets, standardize the features of the training sets and test sets to improve the efficiency and performance of model training, standardize the labels of the training sets, and after prediction, de-standardize the prediction results to obtain the actual values; Select a regression model and use the features and standardized labels on the training set to train two models: one for predicting the amount of electricity purchased corresponding to the electricity price, and the other for predicting the load cost corresponding to the energy storage capacity; use the features on the test set for prediction; For each electricity price in the electricity price sequence P and each energy storage capacity in the energy storage capacity sequence E, the trained model is used to predict the corresponding amount of electricity purchased and load cost; that is, the trained model outputs the amount of electricity purchased corresponding to each electricity price in the electricity price sequence at the future moment And the load cost corresponding to each energy storage capacity in the energy storage capacity sequence at the future moment Combine the prediction results into two data sets: future electricity price-purchased electricity data set and future energy storage capacity-load loss cost dataset 9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the wind power asset transaction management system based on centralized large storage as described in any one of claims 1 to 8.

10. A processor, characterized in that: The processor is used to run a program, wherein when the program is run, the centralized large-scale wind power asset transaction management system according to any one of claims 1 to 8 is executed.

Images