Method for evaluating contribution value of electrochemical energy storage in electric power auxiliary service market

Abstract

The invention discloses a contribution value evaluation method of electrochemical energy storage in an electric power auxiliary service market, and the method comprises the following steps: obtaining first data which comprises an electrochemical energy storage capacity value, power supply data and load energy consumption data, and the power supply data comprises an electric power system power supply value and a renewable energy source integration value; the load energy consumption data comprises an electrochemical energy storage charging value and an electric vehicle charging demand value; calculating a capacity supply coefficient and a capacity consumption coefficient according to the first data; according to the capacity supply coefficient and the capacity consumption coefficient, constructing a contribution value comprehensive evaluation value model of the electrochemical energy storage in the electric power auxiliary service market; and determining the contribution level of the electrochemical energy storage in the electric power auxiliary service market according to the comprehensive evaluation value obtained by the comprehensive evaluation value model. Operation of an electric power auxiliary service market is better guaranteed, and construction of a novel electric power market and a novel electric power system is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of power market management, and in particular to a method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market. Background Art

[0002] The construction of a new power system with a gradually increasing proportion of new energy and a new energy system with the new power system as the core is the primary goal anchored by the current construction and development needs of my country's energy and power industry. With the construction of new power systems, large-scale grid connection of new energy, and growing demand for loads such as electric vehicles, ensuring the effective supply of auxiliary services for new power systems has attracted active exploration by the government and relevant parties. The roles of electrochemical energy storage and traditional power sources such as thermal power in the power auxiliary service market are different, and their contributions and values ​​to the power auxiliary service market are also quite different. Therefore, evaluating the contribution value of electrochemical energy storage in the power auxiliary service market can optimize the resource allocation of the power market, ensure the reasonable sharing of auxiliary service capacity costs among various entities, and give full play to the supply value of stakeholders in the new power system, thereby ensuring the stable power supply and operation of the power system.

[0003] In the existing technology, traditional power market ancillary service management methods usually lack comprehensiveness and are only based on a single indicator or fixed rules to manage power market ancillary services. They are unable to fully consider the complex relationship between power supply data, load energy consumption data and electrochemical energy storage capacity values. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market, which solves the problem that traditional power market auxiliary service management methods are usually lacking in comprehensiveness, only based on a single indicator or fixed rules to manage power market auxiliary services, and unable to fully consider the complex relationship between power supply data, load energy consumption data and electrochemical energy storage capacity value.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market comprises the following steps: obtaining first data, wherein the first data comprises electrochemical energy storage capacity value, power supply data, and load energy consumption data, wherein the power supply data comprises power supply value of the power system and renewable energy integration value, and the load energy consumption data comprises electrochemical energy storage charging value and electric vehicle charging demand value; calculating supply capacity coefficient and capacity consumption coefficient according to the first data; and constructing a comprehensive evaluation value model of the contribution value of electrochemical energy storage in the power auxiliary service market according to the supply capacity coefficient and capacity consumption coefficient, wherein the comprehensive evaluation value model is as follows: Among them, C ois the comprehensive evaluation value of the contribution of electrochemical energy storage in the power auxiliary service market, δ1 is the supply capacity coefficient, δ2 is the capacity consumption coefficient, β1 and β2 are the weight coefficients used to weigh the supply capacity coefficient and the capacity consumption coefficient respectively; the contribution level of electrochemical energy storage in the power auxiliary service market is determined according to the comprehensive evaluation value obtained by the comprehensive evaluation value model.

[0007] Furthermore, the steps for obtaining the supply capacity coefficient δ1 are as follows: the calculation method of the supply capacity coefficient is as follows: a deep learning model is constructed based on the historical power supply data after preprocessing to obtain a power supply data prediction value; the supply capacity coefficient is calculated according to the power supply data prediction value and the electrochemical energy storage capacity value, and the supply capacity coefficient is as follows: Among them, C a is the electrochemical energy storage capacity value, is the predicted value of power supply data, and ε is the power supply capacity modulation factor.

[0008] Furthermore, the historical power supply data includes power supply values ​​of the power system and power output values ​​of renewable energy. The steps for obtaining the predicted power supply data are as follows: construct a deep learning model with a recurrent neural network as the architecture and LSTM as a variant, and use the power supply value of the power system and the power output value of renewable energy as input data, and the predicted power supply data as output data; divide the historical power supply value of the power system and the power output value of renewable energy into a training data set and a verification data set, use the training data set to train the model, and calculate the power supply loss function; use the verification data set to evaluate the model performance, and adjust the model hyperparameters according to the verification results; wherein, the calculation formula of the power supply loss function is as follows:

[0009] in, is the power loss function, i = 1, 2, 3, 4, 5…n, n is the number of training samples, e is a natural constant, y i is the actual power supply data value, is the predicted value of power supply data predicted by the model, is the power loss correction factor.

[0010] Furthermore, the calculation steps of the renewable energy output power value are as follows: obtain the renewable energy integration value and the rated output power of the generator in the power generation system, and calculate the actual proportional coefficient of the generator in the power generation system; standardize the renewable energy integration value and calculate it with the rated output power of the generator in the power generation system to obtain the renewable energy output power value, and the calculation formula is as follows: Among them, E 电 is the output power value of renewable energy, E 标 is the standardized renewable energy integration value, P 额is the rated output power data value of the generator, γ is the actual proportional coefficient, and ρ is the renewable energy modulation factor.

[0011] Furthermore, the specific steps of the standardization of the renewable energy integration value are as follows: obtaining the original data of the renewable energy integration value, and processing the missing values ​​or abnormal values ​​in the original data; converting the processed data into a normal distribution with a mean of 0 and a standard deviation of 1, and the calculation formula is as follows: Among them, Z is the renewable energy integration value, κ is the standard deviation of the original data of renewable energy integration value, X 原 is the original data of renewable energy integration value, φ is the mean of the original data of renewable energy integration value, and ψ is the renewable energy correction factor.

[0012] Furthermore, the calculation steps of the actual proportional coefficient γ are as follows: the historical actual output data values ​​of the generators in the power generation system are collected at the same time interval; the actual proportional coefficient γ is calculated based on the rated output power data of the generator and the historical actual output data values, and the calculation formula is as follows: in, is the historical actual output data value of the generator in the power generation system, d = 1, 2, 3, 4, 5...g, g is the number of calculation groups, P 额 is the rated output power data value of the generator in the power generation system, and θ is the actual proportional counting factor.

[0013] Furthermore, the calculation method of the capacity consumption coefficient δ2 is as follows: a deep learning model is constructed based on the historical load energy consumption data after preprocessing to obtain a load energy consumption data prediction value; the load energy consumption data prediction value and the electrochemical energy storage capacity value are calculated to obtain the capacity consumption coefficient, and the calculation formula is as follows: Among them, C a is the electrochemical energy storage capacity value, is the load energy consumption data predicted by the model, and λ is the capacity consumption modulation factor.

[0014] Furthermore, the historical load energy consumption data includes electrochemical energy storage charging load values ​​and electric vehicle charging demand values. The specific steps for obtaining the load energy consumption data prediction value are as follows: construct a deep learning model based on a convolutional neural network architecture, and use the electrochemical energy storage charging load value and the electric vehicle charging demand value as input data, and the load energy consumption data prediction value as output data; divide the historical electrochemical energy storage charging load value and the electric vehicle charging demand value into a training data set and a validation data set, use the training data set to train the model, and calculate the power consumption loss function; use the validation data set to evaluate the model performance, and adjust the model hyperparameters according to the validation results; wherein, the calculation formula of the power consumption loss function is as follows:

[0015] in, is the power consumption loss function, j = 1, 2, 3, 4, 5…k, k is the number of training samples, z q is the actual power consumption data value, is the predicted value of load energy consumption data predicted by the model, and η is the power loss correction factor.

[0016] Furthermore, the steps for obtaining the electrochemical energy storage charging value are as follows: dividing the annual electrochemical energy storage charging load data into several calculation groups; periodically sampling each calculation group to obtain the electrochemical energy storage charging load observation value at each time point; calculating the sampled electrochemical energy storage charging load observation value to obtain the electrochemical energy storage load value, and the calculation formula is as follows: Where K1 is the electrochemical energy storage charging load value, a = 1, 2, 3, 4, 5 ... b, f = 1, 2, 3, 4, 5 ... t, b is the number of calculation groups, t is the number of sampling points in each calculation group, Q af is the electrochemical energy storage charging load value at the fth time point in the ath calculation group, is the modulation factor in a single group of electrochemical energy storage charging load detection values, and ν is the modulation factor in a single electrochemical energy storage charging load detection value.

[0017] Furthermore, the calculation steps of the electric vehicle charging demand value are as follows: divide the annual electric vehicle charging demand data into several calculation groups, and assign a weight value to each group of electric vehicle charging demand detection values; perform periodic sampling on each calculation group to obtain the electric vehicle charging demand detection value at each time point; screen out the maximum value, minimum value, and abnormal value in the sampled electric vehicle charging demand detection value, and calculate the electric vehicle charging demand value based on the screened data, and the calculation formula is as follows: Where K2 is the charging demand value of electric vehicles, s = 1, 2, 3, 4, 5…u, u = b, v = 1, 2, 3, 4, 5…h, h = t, u is the number of calculation groups, h is the number of sampling time points for each calculation group, P sv is the charging demand value of the electric vehicle at the vth time point in the sth calculation group, α is the modulation factor in the charging demand detection value of a single group of electric vehicles, and l is the counting factor in the charging demand detection value of a single electric vehicle.

[0018] The present invention has the following beneficial effects:

[0019] (1) This method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market provides power market managers with the ability to gain an in-depth understanding of the status of power market auxiliary services by considering multiple key factors such as electrochemical energy storage capacity, power supply data prediction value, load energy consumption data prediction value, supply capacity coefficient, and capacity consumption coefficient, thereby better ensuring the effective supply of market auxiliary service resources and promoting the construction of new power markets and new power systems.

[0020] (2) This method for evaluating the contribution value of electrochemical energy storage in the power ancillary service market introduces factors such as the renewable energy integration value and the generator actual proportion coefficient, and uses clear calculation formulas and steps to make the method more transparent and credible, which helps managers better understand the operation of the power market and provides specific numerical support for the formulation of ancillary service market strategies, thereby helping to promote clean energy integration, reduce dependence on traditional energy, and thus reduce environmental impact;

[0021] (3) The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market improves data quality through grouping and periodic sampling strategies by considering the details of electrochemical energy storage and electric vehicle charging demand, and improves data accuracy through outlier processing, which is crucial for the growing electrochemical energy storage and electric vehicle markets and power market management. In addition, the weighted average method in the method can be customized according to specific needs and situations, providing a more flexible tool for electrochemical energy storage and electric vehicle charging demand analysis;

[0022] (4) The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market takes into account the supply capacity coefficient and the capacity consumption coefficient, and flexibly adjusts the weight coefficients β1 and β2 to meet the capacity management needs under different circumstances. This helps power market managers better understand the balance between the supply capacity and demand capacity of market auxiliary services, and provides key support for the healthy operation of the power market and the stable operation of the power system. In addition, the operability and transparency of the method also make it a powerful tool for power market management.

[0023] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flow chart of a method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to the present invention;

[0025] Figure 2 A flow chart of a method for calculating the supply capacity coefficient in a method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market of the present invention;

[0026] Figure 3 A flow chart of the steps for obtaining the power supply data prediction value in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market according to the present invention;

[0027] Figure 4 A flow chart of the calculation steps of the renewable energy output power value in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market of the present invention;

[0028] Figure 5 A flowchart of specific steps for standardizing the renewable energy integration value in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market according to the present invention;

[0029] Figure 6 A flow chart of the calculation steps of the actual amount ratio coefficient in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market according to the present invention;

[0030] Figure 7 A flow chart of the steps for obtaining the capacity coefficient in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market according to the present invention;

[0031] Figure 8 A flowchart of the specific steps of predicting the load energy consumption data in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market according to the present invention;

[0032] Fig. 9 A flow chart of the steps for obtaining the charging load value of electrochemical energy storage in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market of the present invention;

[0033] Fig.10 This is a flow chart of the calculation steps of the charging demand value of an electric vehicle in a method for evaluating the contribution of electrochemical energy storage in the power auxiliary service market of the present invention. DETAILED DESCRIPTION

[0034] The embodiment of the present application, through a method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market, solves the problem that traditional power market auxiliary service management methods are usually lacking in comprehensiveness, only based on a single indicator or fixed rules to manage power market auxiliary services, and unable to fully consider the complex relationship between power supply data, load energy consumption data and electrochemical energy storage capacity values.

[0035] The overall idea of ​​the problem in the embodiment of this application is as follows:

[0036] By collecting and preprocessing the required data, including load capacity, historical power supply and load energy consumption data, and by comparing the historical power supply data with the load capacity value and the historical load energy consumption data with the load capacity value, the supply capacity coefficient and the capacity consumption coefficient are calculated, and then these coefficients are comprehensively calculated and processed to quantify the comprehensive evaluation value of the contribution of electrochemical energy storage in the power auxiliary service market. This evaluation value will help power market managers better understand the status of auxiliary services to meet changing needs, improve energy utilization efficiency, and promote the integration of renewable energy, thereby ensuring the abundance of auxiliary services in the power market and promoting the healthy development of the power market.

[0037] See also Figure 1 The embodiment of the present invention provides a technical solution: a method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market, comprising the following steps: obtaining first data, the first data including electrochemical energy storage capacity value, power supply data, and load energy consumption data, the power supply data including power supply value of the power system and renewable energy integration value, and the load energy consumption data including electrochemical energy storage charging value and electric vehicle charging demand value; calculating the supply capacity coefficient and capacity consumption coefficient according to the first data; constructing a comprehensive evaluation value model of the contribution value of electrochemical energy storage in the power auxiliary service market according to the supply capacity coefficient and the capacity consumption coefficient, and the comprehensive evaluation value model is as follows: Among them, C o It is a comprehensive evaluation value of the contribution of electrochemical energy storage in the power auxiliary service market, δ1 is the supply capacity coefficient, δ2 is the capacity consumption coefficient, β1 and β2 are weight coefficients used to weigh the supply capacity coefficient and the capacity consumption coefficient respectively.

[0038] The method provides a systematic approach that comprehensively considers the relationship between power supply capacity and load demand in order to more comprehensively evaluate the contribution value of electrochemical energy storage in the power auxiliary service market. By collecting and processing first data, and the first data including electrochemical energy storage capacity value, power supply data, and load energy consumption data, the power supply data includes power system power supply value and renewable energy integration value, and the load energy consumption data includes electrochemical energy storage charging value and electric vehicle charging demand value, the method provides reliable data support for decision makers to help them understand system performance and potential problems. By calculating the supply capacity coefficient and the capacity consumption coefficient, and applying the weight coefficient, the comprehensive evaluation value of the contribution value of electrochemical energy storage in the power auxiliary service market is finally obtained, which makes the evaluation process quantitative and easier to compare with other data. The evaluation results can be used to improve decision making, such as optimizing the supply of auxiliary services, upgrading grid infrastructure, or promoting renewable energy integration to improve the resilience and sustainability of the market; electrochemical storage in power auxiliary The results of the comprehensive evaluation of contribution value in the service market are divided into three situations: high contribution value evaluation value: if the comprehensive evaluation value of contribution value is close to 1, it means that electrochemical energy storage has a good contribution in the power auxiliary service market and has the ability to provide greater auxiliary services to the power market; medium contribution value evaluation value: if the evaluation value is around 0.5, it indicates that electrochemical energy storage has a general contribution in the power auxiliary service market, and its ability to provide auxiliary services to the power market is at a medium level. The market may need some improvements to improve the auxiliary service guarantee capability and enhance the resilience and sustainability of the power market; low contribution value evaluation value: if the evaluation value is close to 0, this may indicate that electrochemical energy storage has a poor contribution in the power auxiliary service market, and its ability to provide auxiliary services to the power market is at a poor level. The power market may have a problem of insufficient auxiliary services, which may require urgent measures to improve the auxiliary service guarantee capability of the power market, such as optimizing user electricity consumption behavior, expanding the types of auxiliary service resources, or improving power supply strategies.

[0039] Specifically, Figure 2 As shown, the steps for obtaining the supply capacity coefficient δ1 are as follows: The calculation method of the supply capacity coefficient is as follows: Based on the historical power supply data after preprocessing, a deep learning model is constructed to obtain the power supply data prediction value; According to the power supply data prediction value and the electrochemical energy storage capacity value, the supply capacity coefficient is calculated, and the supply capacity coefficient is as follows: Among them, C a is the electrochemical energy storage capacity value, is the predicted value of power supply data, and ε is the power supply capacity modulation factor.

[0040] In this implementation scheme, building a deep learning model can improve the accuracy of power supply data, consider historical power supply conditions and other changing factors, and further improve the reliability of the supply capacity coefficient. Secondly, obtaining the predicted value of power supply data is helpful to achieve real-time improvement and timely reflect the power supply situation. By comparing the predicted value of power supply data with the electrochemical energy storage capacity value and calculating the supply capacity coefficient, it can provide strong data support to help decision makers better understand and evaluate the auxiliary service guarantee capability of the power market, ensure the stable operation of the power market under various circumstances, thereby optimizing decision making, improving system performance, and improving the reliability and resilience of the system. The results of the supply capacity coefficient are divided into three cases: the supply capacity coefficient is greater than 0 and less than 1: it means that the supply capacity of the power supply data prediction value exceeds the current electrochemical energy storage capacity value, and there is additional electricity Power supply, in terms of capacity contribution, represents the excess capacity of the power supply data forecast value, which can provide effective auxiliary services for the market, and is a positive trend; the supply capacity coefficient is equal to 1: it means that the supply capacity of the power supply data forecast value is balanced with the current electrochemical energy storage capacity value electricity demand, and there is no excess or shortage of electricity. In terms of capacity contribution, it means that the power supply data forecast value can effectively meet the electrochemical energy storage capacity value electricity demand at present, which is a balanced state; the supply capacity coefficient is greater than 1: it means that the supply capacity of the power supply data forecast value is insufficient to meet the current electrochemical energy storage capacity value electricity demand, and measures need to be taken to balance the electrochemical energy storage charging load capacity value electricity demand. In terms of capacity contribution, it means that the electrochemical energy storage charging load capacity value electricity demand exceeds the supply capacity of the power supply data forecast value, which is a negative trend.

[0041] Specifically, Figure 3 As shown in the figure, the historical power supply data includes the power supply value of the power system and the output power value of renewable energy. The steps for obtaining the predicted value of the power supply data are as follows: construct a deep learning model with a recurrent neural network as the architecture and LSTM as a variant, and use the power supply value of the power system and the output power value of renewable energy as input data, and the predicted value of the power supply data as output data; divide the historical power supply value of the power system and the output power value of renewable energy into a training data set and a validation data set, use the training data set to train the model, and calculate the power supply loss function; use the validation data set to evaluate the model performance, and adjust the model hyperparameters according to the validation results; the calculation formula of the power supply loss function is as follows:

[0042] in, is the power loss function,

[0043] i = 1, 2, 3, 4, 5…n, n is the number of training samples, e is a natural constant, y i is the actual power supply data value, is the predicted value of power supply data predicted by the model, is the power loss correction factor.

[0044] In this implementation scheme, the power supply data prediction based on the deep learning model can provide more accurate estimation of future power supply data. The recurrent neural network architecture such as LSTM has excellent modeling ability for time series data, which can better capture the time series relationship between the power supply value of the power system, the output value of renewable energy, and the power supply data prediction value, thereby improving the prediction accuracy. This method can quickly update the prediction of power supply data to adapt to actual needs and the real-time requirements of the system. This enables decision makers to better respond to sudden changes in power demand. By using a validation data set to evaluate model performance and adjusting the model's hyperparameters based on the validation results, the generalization performance of the model can be improved to adapt to different data and scenarios. The definition of the power loss function can evaluate the performance of the model. By adjusting the hyperparameters of its model to minimize the power loss function, the model's prediction can be made more accurate, thereby ensuring the quality of power supply data. This enables system managers to better predict power demand, make accurate decisions, optimize power supply, and improve system performance.

[0045] Specifically, Figure 4 As shown, the calculation steps of the renewable energy output power value are as follows: obtain the renewable energy integration value and the rated output power of the generator in the power generation system, and calculate the actual proportional coefficient of the generator in the power generation system; standardize the renewable energy integration value and calculate it with the rated output power of the generator in the power generation system to obtain the renewable energy output power value, and the calculation formula is as follows: Among them, E 电 is the output power value of renewable energy, E 标 is the standardized renewable energy integration value, P 额 is the rated output power data value of the generator, γ is the actual proportional coefficient, and ρ is the renewable energy modulation factor.

[0046] In this implementation scheme, this method comprehensively considers multiple important factors, including the integration level of renewable energy, the rated output power of the generator, the actual proportion coefficient and standardized processing. This comprehensive consideration can more accurately reflect the actual output of renewable energy, and can comprehensively evaluate the integration level of renewable energy in the power system, helping decision makers understand the contribution of renewable energy to the system. By combining these factors, the method can be adapted according to the actual situation in the power market to ensure that the calculated renewable energy output power value is consistent with the actual situation, so as to better reflect the actual output of renewable energy to meet real-time power demand, thereby making the market more flexible and adaptable. Since there is a loss error between the actual output power and the rated output power of the generator in the power generation system, the accuracy of renewable energy output can be improved by using the rated output power data and the actual proportion coefficient. Providing accurate renewable energy output power values ​​will help decision makers better plan the power system, formulate policies and allocate resources to promote the integration of renewable energy and improve the sustainability of the power system.

[0047] Specifically, Figure 5 As shown in the figure, the specific steps of standardization of renewable energy integration value are as follows: obtain the original data of renewable energy integration value, and process the missing values ​​or abnormal values ​​in the original data; convert the processed data into a normal distribution with a mean of 0 and a standard deviation of 1, and the calculation formula is as follows: Among them, Z is the renewable energy integration value, κ is the standard deviation of the original data of renewable energy integration value, X 原 is the original data of renewable energy integration value, φ is the mean of the original data of renewable energy integration value, and ψ is the renewable energy correction factor.

[0048] In this implementation scheme, since the raw data of renewable energy integration values ​​usually have different scales and ranges, this may make the data difficult to compare or cause the model to be unstable, and the normal distribution is usually an ideal state of data distribution, which meets the prerequisites of many statistical models. By standardizing them with a normal distribution with a mean of 0 and a standard deviation of 1, the data are scaled to a uniform scale so that they can be effectively compared, which makes it easier to compare and analyze renewable energy integration values ​​in different time periods or locations. In machine learning and statistical modeling, models are usually more stable for standardized data, easier to train and converge, so that the standardization of renewable energy integration values ​​helps to improve the performance and reliability of the model.

[0049] Specifically, Figure 6As shown, the calculation steps of the actual proportional coefficient γ are as follows: the historical actual output data values ​​of the generators in the power generation system are collected at the same time interval; the actual proportional coefficient γ is calculated based on the rated output power data of the generator and the historical actual output data values, and the calculation formula is as follows: in, is the historical actual output data value of the generator in the power generation system, d = 1, 2, 3, 4, 5...g, g is the number of calculation groups, P 额 is the rated output power data value of the generator in the power generation system, and θ is the actual proportional counting factor.

[0050] In this implementation scheme, the complexity of this actual proportional coefficient calculation method lies in that it uses multiple historical actual output data values, rather than relying solely on theoretical values. Such complexity brings higher authenticity and adaptability, because it better considers the actual working conditions and performance changes of generators in the power system. By collecting multiple historical data points, it more comprehensively considers the influence of factors such as seasonality, weather conditions, and load fluctuations, making the calculation of the actual proportional coefficient more accurate and reliable. Such accuracy is crucial to supporting decision-making and performance optimization of the power system. Therefore, this actual proportional coefficient calculation method will help decision makers better understand the actual performance of the power system and provide them with reliable data support to cope with changing power demand and environmental conditions and improve the sustainability and efficiency of the system.

[0051] Specifically, Figure 7 As shown, the calculation method of the capacity consumption coefficient δ2 is as follows: based on the historical load energy consumption data after preprocessing, a deep learning model is constructed to obtain the load energy consumption data prediction value; the load energy consumption data prediction value and the electrochemical energy storage capacity value are calculated to obtain the capacity consumption coefficient, and the calculation formula is as follows:

[0052] Among them, C a is the electrochemical energy storage capacity value, is the load energy consumption data predicted by the model, and λ is the capacity consumption modulation factor.

[0053] In this implementation scheme, the use of the deep learning model in this step improves the prediction accuracy of the load energy consumption data. This is because the deep learning model can better capture the potential patterns and complex relationships in the historical data, and therefore can provide more reliable prediction results. At the same time, through multiple predictions, the system can better adapt to the changing power demand, so as to improve the real-time and adaptability of the system. By calculating the capacity coefficient, the system manager can better understand the demand and supply of auxiliary services in the power market. This helps to make more informed decisions, such as optimizing power supply, rationally planning resource allocation, and improving system performance. This combination of complexity and data support can provide power market information, help decision makers better understand the market operation, so that they can better plan and manage the power market, ensure that the market can cope with load fluctuations and changes, and improve the resilience and sustainability of the market. The results of the supply capacity coefficient are divided into three situations: the supply capacity coefficient is greater than 1: it means that the supply capacity of the power supply data prediction value is insufficient to meet the current electrochemical energy storage charging load capacity value electricity demand, and measures need to be taken to balance the electrochemical energy storage charging load capacity value electricity demand. In terms of contribution, it represents that the electrochemical energy storage charging load capacity value electricity demand exceeds the supply capacity of the power supply data prediction value. , which is a negative trend; the supply capacity coefficient is equal to 1: it means that the supply capacity of the power supply data forecast value is balanced with the current electrochemical energy storage charging load capacity value electricity demand, and there is no excess or shortage of electricity. In terms of contribution, it means that the power supply data forecast value can effectively meet the electrochemical energy storage charging load capacity value electricity demand, which is a balanced state; the supply capacity coefficient is greater than 0 and less than 1: it means that the supply capacity of the power supply data forecast value exceeds the current electrochemical energy storage charging load capacity value electricity demand, and there is additional power supply. In terms of contribution, it represents the excess auxiliary service capacity of the power supply data forecast value, which can provide effective auxiliary services to the market, which is a positive trend.

[0054] Specifically, Figure 8 As shown in the figure, the historical load energy consumption data includes the electrochemical energy storage charging load value and the electric vehicle charging demand value. The specific steps to obtain the load energy consumption data prediction value are as follows: construct a deep learning model with a convolutional neural network as the architecture, and use the electrochemical energy storage charging load value and the electric vehicle charging demand value as input data, and the load energy consumption data prediction value as output data; divide the historical electrochemical energy storage charging load value and the electric vehicle charging demand value into a training data set and a validation data set, use the training data set to train the model, and calculate the power consumption loss function; use the validation data set to evaluate the model performance, and adjust the model hyperparameters according to the validation results; the calculation formula of the power consumption loss function is as follows: in, is the power consumption loss function, j = 1, 2, 3, 4, 5…k, k is the number of training samples, z qis the actual power consumption data value, is the predicted value of load energy consumption data predicted by the model, and η is the power loss correction factor.

[0055] In this implementation, with the help of a convolutional neural network deep learning model, combined with the electrochemical energy storage charging load value and the electric vehicle charging demand value, future load energy consumption can be predicted more accurately. The deep learning model can automatically capture the features and patterns in the data, thereby providing high-precision prediction results, which helps to better understand and respond to load fluctuations. By continuously using training data to optimize the model, defining the power consumption loss function and minimizing it, and using verification data to evaluate performance, real-time data prediction and model updates can be achieved. This is crucial for real-time adjustment and decision-making of the power system. By predicting load energy consumption data, decision makers can have a more comprehensive understanding of the demand and supply of the power system, which helps to make more informed decisions, such as optimization of power supply and resource allocation.

[0056] Specifically, Fig. 9 As shown, the steps for obtaining the electrochemical energy storage charging load value are as follows: divide the annual electrochemical energy storage charging load data into several calculation groups; perform periodic sampling on each calculation group to obtain the electrochemical energy storage charging load observation value at each time point; calculate the sampled electrochemical energy storage charging load observation value to obtain the electrochemical energy storage load value, and the calculation formula is as follows: Where K1 is the electrochemical energy storage load value, a = 1, 2, 3, 4, 5 ... b, f = 1, 2, 3, 4, 5 ... t, b is the number of calculation groups, t is the number of sampling points in each calculation group, Q af is the electrochemical energy storage charging load value at the fth time point in the ath calculation group, is the modulation factor in a single group of electrochemical energy storage charging load detection values, and ν is the modulation factor in a single electrochemical energy storage charging load detection value.

[0057] In this embodiment, the step of acquiring the electrochemical energy storage charging load value fully considers the complexity of time and multi-dimensional data. First, by dividing the annual data into calculation groups and performing periodic sampling, it can more accurately reflect the seasonal and periodic changes in the power load, thereby improving the timing and accuracy of the data; secondly, the modulation factor is introduced to make the data more diverse, which helps the model to better capture the differences in power load between different groups and time points; most importantly, this method improves data quality through refined data processing, provides more reliable data support for the power system, helps to improve system performance, refine resource planning and decision-making, and ensure that electricity demand is better met. Therefore, the acquisition step provides more advantages for the analysis and management of the power system by comprehensively considering the timing and multi-dimensional characteristics of the data, and provides strong support for sustainable power supply.

[0058] Specifically, Fig.10 As shown in the figure, the calculation steps of the electric vehicle charging demand value are as follows: divide the annual electric vehicle charging demand data into several calculation groups, and assign a weight value to each group of electric vehicle charging demand detection values; perform periodic sampling on each calculation group to obtain the electric vehicle charging demand detection value at each time point; filter out the maximum value, minimum value, and abnormal value in the sampled electric vehicle charging demand detection value, and calculate the electric vehicle charging demand value based on the filtered data. The calculation formula is as follows: Where K2 is the charging demand value of electric vehicles, s = 1, 2, 3, 4, 5…u, u = b, v = 1, 2, 3, 4, 5…h, h = t, u is the number of calculation groups, h is the number of sampling time points for each calculation group, P sv is the charging demand value of the electric vehicle at the vth time point in the sth calculation group, α is the modulation factor in the charging demand detection value of a single group of electric vehicles, and l is the counting factor in the charging demand detection value of a single electric vehicle.

[0059] In this implementation, the calculation steps of the electric vehicle charging demand value firstly take into account the differences between different data groups through grouping and weighting, so as to more accurately capture the diversity of data, thereby improving the depth and breadth of data analysis. Secondly, by screening out abnormal data and unreasonable maximum and minimum values, the method ensures the credibility and accuracy of the calculated charging demand data. This helps to eliminate abnormal situations that may interfere with the analysis of the power system and make the data more reliable. At the same time, the calculation formula The use of modulation factors and counting factors enables more refined data processing and further enhances the quality of data. This refined processing helps the model better adapt to actual needs, improves data diversity, and provides more comprehensive support for power system analysis and decision-making.

[0060] In summary, this application has at least the following effects:

[0061] By considering multiple key factors such as electrochemical energy storage capacity, power supply data forecast, load energy consumption data forecast, supply capacity coefficient, and capacity consumption coefficient, it provides power market managers with the ability to gain an in-depth understanding of the status of power market ancillary services, thereby better ensuring the effective supply of market ancillary services and promoting the construction of new power markets and new power systems.

[0062] By introducing factors such as renewable energy integration value and generator actual ratio coefficient, and through clear calculation formulas and steps, the method is made more transparent and credible, which helps managers better understand the operation of the power system and power market, and provides specific numerical support for the formulation of market strategies, thereby helping to promote clean energy integration, reduce dependence on traditional energy, and thus reduce environmental impact.

[0063] By considering the details of EV charging demand, data quality is improved through strategies such as grouping and periodic sampling, and data accuracy is improved through outlier processing, which is crucial for the growing EV market and electricity market management. In addition, the weighted average method in the method can be customized according to specific needs and situations, providing a more flexible tool for EV charging demand analysis.

[0064] This method comprehensively considers the supply capacity coefficient and the capacity consumption coefficient, and meets the management needs in different situations through flexible adjustment of the weight coefficients β1 and β2. This helps power market managers better understand the balance between capacity supply capacity and demand capacity, and provides key support for the stable operation of the power market. In addition, the operability and transparency of the method also make it a powerful tool for power market management and planning.

[0065] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market, characterized in that: The following steps are involved: Acquire first data, wherein the first data includes an electrochemical energy storage capacity value, power supply data, and load energy consumption data, wherein the power supply data includes a power supply value of a power system and a renewable energy integration value, and the load energy consumption data includes an electrochemical energy storage charging value and an electric vehicle charging demand value; Calculate the supply capacity coefficient and the capacity consumption coefficient according to the first data; According to the supply capacity coefficient and capacity consumption coefficient, a comprehensive evaluation model of the contribution value of electrochemical energy storage in the power auxiliary service market is constructed. The comprehensive evaluation model is as follows: Among them, C o is the comprehensive evaluation value of the contribution of electrochemical energy storage in the power auxiliary service market, δ1 is the supply capacity coefficient, δ2 is the capacity consumption coefficient, β1 and β2 are the weight coefficients used to weigh the supply capacity coefficient and the capacity consumption coefficient respectively; Based on the comprehensive evaluation value obtained from the comprehensive evaluation value model, the contribution level of electrochemical energy storage in the power auxiliary service market is determined.

2. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 1 is characterized in that: The calculation method of the supply capacity coefficient is as follows: A deep learning model is built based on the pre-processed historical power supply data to obtain the power supply data prediction value; According to the power supply data prediction value and the electrochemical energy storage capacity value, the supply capacity coefficient is calculated, and the supply capacity coefficient is as follows: Among them, C a is the electrochemical energy storage capacity value, is the predicted value of power supply data, ε is the power supply capacity modulation factor, and e is a natural constant.

3. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 2 is characterized in that: The historical power supply data includes the power supply value of the power system and the output power value of renewable energy. The steps for obtaining the power supply data prediction value are as follows: Build a deep learning model with a recurrent neural network as the architecture and LSTM as a variant, with the power supply value of the power system and the output value of renewable energy as input data, and the power supply data prediction value as output data; The historical power supply values ​​of the power system and the output power values ​​of renewable energy are divided into a training data set and a validation data set, the training data set is used to train the model, and the power supply loss function is calculated; Use the validation dataset to evaluate model performance and adjust model hyperparameters based on the validation results; The calculation formula of the power supply loss function is as follows: Among them, θ1 is the power loss function, i = 1, 2, 3, 4, 5…n, n is the number of training samples, e is a natural constant, y i is the actual power supply data value, is the predicted value of power supply data predicted by the model, is the power loss correction factor.

4. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 3 is characterized in that: The calculation steps for the renewable energy output value are as follows: Obtain the renewable energy integration value and the rated output power of the generator in the power generation system, and calculate the actual proportional coefficient of the generator in the power generation system; The renewable energy integration value is standardized and calculated with the rated output power of the generator in the power generation system to obtain the renewable energy output power value. The calculation formula is as follows: Among them, E 电 is the output power value of renewable energy, E 标 is the standardized renewable energy integration value, P 额 is the rated output power data value of the generator, γ is the actual proportional coefficient, and ρ is the renewable energy modulation factor.

5. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 4 is characterized in that: The specific steps of the standardization process of the renewable energy integration value are as follows: Obtain the original data of renewable energy integration value and process the missing values ​​or abnormal values ​​in the original data; The processed data is converted into a normal distribution with a mean of 0 and a standard deviation of 1. The calculation formula is as follows: Among them, Z is the renewable energy integration value, κ is the standard deviation of the original data of renewable energy integration value, X 原 is the original data of renewable energy integration value, φ is the mean of the original data of renewable energy integration value, and ψ is the renewable energy correction factor.

6. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 4 is characterized in that: The calculation steps of the actual amount ratio coefficient γ are as follows: Collecting historical actual output data values ​​of generators in a plurality of power generation systems at the same time intervals; The actual proportional coefficient γ is calculated based on the rated output power data of the generator and several historical actual output data values. The calculation formula is as follows: in, is the historical actual output data value of the generator in the power generation system, d = 1, 2, 3, 4, 5...g, g is the number of calculation groups, P 额 is the rated output power data value of the generator in the power generation system, and θ is the actual proportional counting factor.

7. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 1 is characterized in that: The calculation method of the capacity loss coefficient δ2 is as follows: Based on the pre-processed historical load energy consumption data, a deep learning model is constructed to obtain the load energy consumption data prediction value; The load energy consumption data prediction value and the electrochemical energy storage capacity value are used to calculate the capacity consumption coefficient, and the calculation formula is as follows: Among them, C a is the electrochemical energy storage capacity value, is the load energy consumption data predicted by the model, and λ is the capacity consumption modulation factor.

8. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 7 is characterized in that: The historical load energy consumption data includes the electrochemical energy storage charging value and the electric vehicle charging demand value. The specific steps for obtaining the load energy consumption data prediction value are as follows: Construct a deep learning model based on convolutional neural network architecture, with electrochemical energy storage charging value and electric vehicle charging demand value as input data, and load energy consumption data prediction value as output data; The historical electrochemical energy storage charging values ​​and electric vehicle charging demand values ​​are divided into a training data set and a validation data set, the training data set is used to train the model, and the power consumption loss function is calculated; Use the validation dataset to evaluate model performance and adjust model hyperparameters based on the validation results; Among them, the calculation formula of the power consumption loss function is as follows: Among them, θ2 is the power consumption loss function, j = 1, 2, 3, 4, 5…k, k is the number of training samples, z q is the actual power consumption data value, is the predicted value of load energy consumption data predicted by the model, and η is the power loss correction factor.

9. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 8 is characterized in that: The steps for obtaining the electrochemical energy storage charging value are as follows: Divide the annual electrochemical energy storage charging load data into several calculation groups; Perform periodic sampling on each calculation group to obtain the electrochemical energy storage charging load observation value at each time point; The sampled electrochemical energy storage charging load observation value is calculated to obtain the electrochemical energy storage charging value, and the calculation formula is as follows: Where K1 is the electrochemical energy storage charge value, a = 1, 2, 3, 4, 5 ... b, f = 1, 2, 3, 4, 5 ... t, b is the number of calculation groups, t is the number of sampling points in each calculation group, Q af is the electrochemical energy storage charging load value at the fth time point in the ath calculation group, is the modulation factor in a single group of electrochemical energy storage charging load detection values, and ν is the modulation factor in a single electrochemical energy storage charging load detection value.

10. The method for evaluating the contribution value of electrochemical energy storage in the power auxiliary service market according to claim 8, characterized in that: The calculation steps of the electric vehicle charging demand value are as follows: Divide the annual electric vehicle charging demand data into several calculation groups, and assign a weight value to each group of electric vehicle charging demand detection values; Perform periodic sampling on each calculation group to obtain the electric vehicle charging demand detection value at each time point; The maximum value, minimum value and abnormal value in the sampled electric vehicle charging demand detection value are screened out, and the electric vehicle charging demand value is calculated based on the screened data. The calculation formula is as follows: Where K2 is the charging demand value of electric vehicles, s = 1, 2, 3, 4, 5…u, u = b, v = 1, 2, 3, 4, 5…h, h = t, u is the number of calculation groups, h is the number of sampling time points for each calculation group, P sv is the charging demand value of the electric vehicle at the vth time point in the sth calculation group, α is the modulation factor in the charging demand detection value of a single group of electric vehicles, and l is the counting factor in the charging demand detection value of a single electric vehicle.

Images