A charging and discharging optimization method and device for energy storage power station

Abstract

The embodiment of the present application provides a charging and discharging optimization method and device for an energy storage power station, which relates to the technical field of industrial and commercial energy storage power stations. The method includes: obtaining historical data related to charging and discharging of the energy storage power station, including historical load data, holiday information, electricity price information, and historical weather information; constructing multiple load prediction models, and using the historical data related to charging and discharging to train each load prediction model to obtain multiple trained load prediction models, and selecting the optimal load prediction model from the multiple trained load prediction models; using the optimal load prediction model to predict the predicted load result of the energy storage power station within a preset time period in the future, and using the predicted load result of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of industrial and commercial energy storage power stations, and in particular to a method and device for optimizing charging and discharging of an energy storage power station. Background Art

[0002] The main profit model of industrial and commercial energy storage power stations is peak shaving and valley filling, that is, charging and storing electricity when electricity demand is low (low electricity prices), and discharging and supplying energy when electricity demand is high (high electricity prices). Load forecasting can predict the electricity demand of users in the future, thereby guiding energy storage power stations to formulate scientific operation strategies and improve the economic and social benefits of energy storage power stations.

[0003] Common load forecasting methods include classical methods (time series method, similar day method, regression analysis method), modern algorithms (neural networks, fuzzy mathematics theory, support vector machines), and combined algorithms (combination of time series method and neural network, etc.). Due to the complexity of load influencing factors, such as weather, economy, social activities, etc., there are complex interactions between these factors, making it difficult for the forecasting model to accurately capture their impact. In addition, load forecasting has obvious seasonal and periodic characteristics, and may also be affected by sudden events, which requires the forecasting model to have high adaptability and be able to flexibly adjust the forecasting strategy according to different situations. Due to the randomness and uncertainty of load changes, the forecast results often have certain volatility and errors. How to improve the stability and reliability of the forecast results is an important issue that needs to be solved in load forecasting.

[0004] The charging and discharging strategy of energy storage power stations significantly affects the operating income of power stations. Taking a commercial and industrial energy storage power station as an example, due to the existence of two arbitrage modes, valley charging and peak discharging and flat charging and peak discharging, the traditional energy storage operation mode is to perform two full charges (both valley charging and flat charging are fully charged to 100% SOC). The disadvantage of this charging and discharging strategy is that it ignores the insufficient absorption capacity during the evening peak, which will cause excessive charging during the normal period, affecting the charging during the valley period of the next day, resulting in increased operating costs and reduced income of the power station.

[0005] Therefore, how to accurately match power demand with the charging and discharging capacity of energy storage power stations through accurate load forecasting, help energy storage power stations charge and discharge at the best time, and maximize the benefits of energy storage power stations is extremely practical. Summary of the invention

[0006] In view of this, an embodiment of the present application provides a method for optimizing charging and discharging of an energy storage power station. One or more embodiments of the present application also involve a charging and discharging optimization device for an energy storage power station, a computing device, a computer-readable storage medium, and a computer program to solve the technical defects existing in the prior art.

[0007] According to a first aspect of an embodiment of the present application, a method for optimizing charging and discharging of an energy storage power station is provided, comprising:

[0008] Obtain historical data related to charging and discharging of energy storage power stations, including historical load data, holiday information, electricity price information, and historical weather information;

[0009] Constructing multiple load prediction models, and using the historical data related to charging and discharging to train each load prediction model to obtain multiple trained load prediction models, and selecting the optimal load prediction model from the multiple trained load prediction models;

[0010] The optimal load forecasting model is used to predict the predicted load results of the energy storage power station within a preset time period in the future, and the predicted load results of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python are used to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

[0011] Preferably, the multiple load forecasting models include: a load forecasting model of a time series network structure, a load forecasting model of a tree network structure, and a load forecasting model of a combined network structure including a time series network structure and a tree network structure.

[0012] Preferably, the use of the historical data related to charging and discharging to train each load forecasting model to obtain a plurality of trained load forecasting models includes:

[0013] Using the historical load data and holiday information in the historical data related to charging and discharging to train the load forecasting model of the time series network structure, to obtain a trained load forecasting model of the time series network structure;

[0014] Using the historical load data, holiday information, electricity price information and historical weather information in the historical data related to charging and discharging to train the load forecasting model of the tree network structure, to obtain a trained load forecasting model of the tree network structure;

[0015] The load forecasting model of the combined network structure is trained using the historical load data, holiday information, electricity price information and historical weather information in the historical data related to charging and discharging, so as to obtain a trained load forecasting model of the combined network structure.

[0016] Preferably, the selecting the optimal load forecasting model from the trained multiple load forecasting models comprises:

[0017] Calculate in real time a first real-time prediction accuracy of the load prediction model of the time series network structure, and calculate a first average prediction accuracy of the load prediction model of the time series network structure based on the first real-time prediction accuracy;

[0018] Calculating in real time a second real-time prediction accuracy of the load prediction model of the tree network structure, and calculating a second average prediction accuracy of the load prediction model of the tree network structure according to the second real-time prediction accuracy;

[0019] Calculate in real time a third real-time prediction accuracy of the load prediction model of the combined network structure, and calculate a third average prediction accuracy of the load prediction model of the combined network structure based on the third real-time prediction accuracy;

[0020] According to the first average prediction accuracy, the second average prediction accuracy and the third average prediction accuracy, a load prediction model with the largest average prediction accuracy is selected as the optimal load prediction model.

[0021] Preferably, the method of using the predicted load result of the energy storage power station in a preset time period in the future and the charge and discharge power optimization model constructed by the pyscipopt module of python to obtain the optimal charge and discharge power of the energy storage power station at each moment in the future includes:

[0022] Obtaining basic information of the energy storage power station including transformer capacity, energy storage power, energy storage capacity and electricity price information;

[0023] Taking maximizing the charging and discharging benefits of the energy storage power station as the goal, constructing an objective function;

[0024] Construct constraints including energy storage SOC constraints, charging power constraints, discharging power constraints and transformer capacity constraints;

[0025] Constructing a charge-discharge power optimization model through the pyscipopt module of python, and adding the objective function and the constraint condition to the charge-discharge power optimization model to obtain a target charge-discharge power optimization model;

[0026] By inputting the basic information of the energy storage power station including transformer capacity, energy storage power, energy storage capacity and electricity price information and the predicted load results of the energy storage power station in a future preset time period into the target charging and discharging power optimization model, the optimal charging and discharging power of the energy storage power station at each moment in the future is obtained.

[0027] Preferably, the constraints include:

[0028] The energy storage SOC constraint is that the SOC at any time is between 0 and 1;

[0029] The charging power constraint is that the charging power at any time is between 0 and the maximum charging power;

[0030] The discharge power constraint is that the discharge power at any time is between 0 and the maximum discharge power and the discharge power at any time is not greater than the load power;

[0031] The transformer capacity constraint is that the transformer capacity at any time is greater than or equal to the sum of the load power and the charging power.

[0032] According to a second aspect of an embodiment of the present application, a charging and discharging optimization device for an energy storage power station is provided, comprising:

[0033] An acquisition module is configured to acquire historical data related to charging and discharging of the energy storage power station including historical load data, holiday information, electricity price information and historical weather information;

[0034] A model optimization module is configured to construct multiple load prediction models, and train each load prediction model using the historical data related to charging and discharging to obtain multiple trained load prediction models, and select the optimal load prediction model from the multiple trained load prediction models;

[0035] The charging and discharging optimization module is configured to use the optimal load prediction model to predict the predicted load results of the energy storage power station within a preset time period in the future, and use the predicted load results of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

[0036] Preferably, the multiple load forecasting models include: a load forecasting model of a time series network structure, a load forecasting model of a tree network structure, and a load forecasting model of a combined network structure including a time series network structure and a tree network structure.

[0037] According to a third aspect of an embodiment of the present application, a computing device is provided, including:

[0038] Memory and processor;

[0039] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement any step of the method for optimizing charging and discharging of the energy storage power station.

[0040] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of any one of the methods for optimizing charging and discharging of an energy storage power station are implemented.

[0041] According to a fifth aspect of an embodiment of the present application, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned method for optimizing charging and discharging of an energy storage power station.

[0042] The charging and discharging optimization scheme for the energy storage power station provided in the embodiment of the present application is obtained by obtaining the historical data related to charging and discharging of the energy storage power station including historical load data, holiday information, electricity price information and historical weather information; constructing multiple load forecasting models, and using the historical data related to charging and discharging to train each load forecasting model to obtain multiple trained load forecasting models, and selecting the optimal load forecasting model from the multiple trained load forecasting models; using the optimal load forecasting model to predict the predicted load results of the energy storage power station in the future preset time period, and using the predicted load results of the energy storage power station in the future preset time period and the charging and discharging power optimization model constructed by the pyscipopt module of python, to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future. The charging and discharging optimization scheme for the energy storage power station provided in the embodiment of the present invention reduces the user's electricity cost, improves the operating income of the energy storage power station, and better serves the operation and development of industrial and commercial energy storage power stations while improving the accuracy and reliability of the power station load forecast. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of a charging and discharging optimization method of an energy storage power station provided by an embodiment of the present application;

[0044] Figure 2 is a schematic diagram of a charging and discharging optimization device of an energy storage power station provided by an embodiment of the present application;

[0045] Figure 3 It is an overall flow chart of a charging and discharging optimization method of an energy storage power station provided by an embodiment of the present application;

[0046] Figure 4 This is a schematic diagram of a charge and discharge strategy based on load prediction provided by an embodiment of the present application;

[0047] Figure 5 This is a schematic diagram of improving charging and discharging benefits based on load prediction provided by an embodiment of the present application;

[0048] Figure 6 It is a structural block diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0049] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0050] The terms used in one or more embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present application. The singular forms of "a", "said" and "the" used in one or more embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present application refers to and includes any or all possible combinations of one or more associated listed items.

[0051] It should be understood that, although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present application, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0052] In the present application, a method for optimizing charging and discharging of an energy storage power station is provided. The present application also relates to a charging and discharging optimization device for an energy storage power station, a computing device, a computer-readable storage medium, and a computer program, which are described in detail one by one in the following embodiments.

[0053] Figure 1 A flow chart of a method for optimizing charging and discharging of an energy storage power station provided according to an embodiment of the present application is shown, which specifically includes the following steps.

[0054] Step 101: Acquire historical data related to charging and discharging of the energy storage power station including historical load data, holiday information, electricity price information and historical weather information;

[0055] Step 102: construct multiple load prediction models, and use the historical data related to charging and discharging to train each load prediction model to obtain multiple trained load prediction models, and select the optimal load prediction model from the multiple trained load prediction models;

[0056] In one embodiment of the present application, the multiple load forecasting models include: a load forecasting model of a time series network structure, a load forecasting model of a tree network structure, and a load forecasting model of a combined network structure including a time series network structure and a tree network structure.

[0057] In one embodiment of the present application, the use of the historical data related to charging and discharging to train each load forecasting model to obtain multiple trained load forecasting models includes:

[0058] Using the historical load data and holiday information in the historical data related to charging and discharging to train the load forecasting model of the time series network structure, to obtain a trained load forecasting model of the time series network structure;

[0059] Using the historical load data, holiday information, electricity price information and historical weather information in the historical data related to charging and discharging to train the load forecasting model of the tree network structure, to obtain a trained load forecasting model of the tree network structure;

[0060] The load forecasting model of the combined network structure is trained using the historical load data, holiday information, electricity price information and historical weather information in the historical data related to charging and discharging, so as to obtain a trained load forecasting model of the combined network structure.

[0061] In one embodiment of the present application, selecting the optimal load forecasting model from the trained multiple load forecasting models includes:

[0062] Calculate in real time a first real-time prediction accuracy of the load prediction model of the time series network structure, and calculate a first average prediction accuracy of the load prediction model of the time series network structure based on the first real-time prediction accuracy;

[0063] Calculating in real time a second real-time prediction accuracy of the load prediction model of the tree network structure, and calculating a second average prediction accuracy of the load prediction model of the tree network structure according to the second real-time prediction accuracy;

[0064] Calculate in real time a third real-time prediction accuracy of the load prediction model of the combined network structure, and calculate a third average prediction accuracy of the load prediction model of the combined network structure based on the third real-time prediction accuracy;

[0065] According to the first average prediction accuracy, the second average prediction accuracy and the third average prediction accuracy, a load prediction model with the largest average prediction accuracy is selected as the optimal load prediction model.

[0066] Step 103: Use the optimal load forecasting model to predict the predicted load results of the energy storage power station within a preset time period in the future, and use the predicted load results of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

[0067] In one embodiment of the present application, the optimal charging and discharging power of the energy storage power station at each moment in the future is obtained by using the predicted load result of the energy storage power station in the future preset time period and the charging and discharging power optimization model constructed by the pyscipopt module of python, including:

[0068] Obtaining basic information of the energy storage power station including transformer capacity, energy storage power, energy storage capacity and electricity price information;

[0069] Taking maximizing the charging and discharging benefits of the energy storage power station as the goal, constructing an objective function;

[0070] Construct constraints including energy storage SOC constraints, charging power constraints, discharging power constraints and transformer capacity constraints;

[0071] Constructing a charge-discharge power optimization model through the pyscipopt module of python, and adding the objective function and the constraint condition to the charge-discharge power optimization model to obtain a target charge-discharge power optimization model;

[0072] By inputting the basic information of the energy storage power station including transformer capacity, energy storage power, energy storage capacity and electricity price information and the predicted load results of the energy storage power station in a future preset time period into the target charging and discharging power optimization model, the optimal charging and discharging power of the energy storage power station at each moment in the future is obtained.

[0073] In one embodiment of the present application, the constraints include:

[0074] The energy storage SOC constraint is that the SOC at any time is between 0 and 1;

[0075] The charging power constraint is that the charging power at any time is between 0 and the maximum charging power;

[0076] The discharge power constraint is that the discharge power at any time is between 0 and the maximum discharge power and the discharge power at any time is not greater than the load power;

[0077] The transformer capacity constraint is that the transformer capacity at any time is greater than or equal to the sum of the load power and the charging power.

[0078] In summary, the embodiment of the present application includes two parts:

[0079] Part I: Load forecasting method based on multi-model automatic optimization

[0080] Based on the power load characteristics of different users, a set of load forecasting models (including time series models (such as prophet), tree models (such as xgboost), combined models (such as prophet+xgboost), etc.) are built to make rolling forecasts of the power load of users in the future (such as 48 hours). And based on the measured load data, the load forecast accuracy is calculated in real time. By calculating the load forecast accuracy of each model for a period of time in the history of the station (such as 3 days in history), the most accurate forecast model is determined, and its forecast result is used as the load forecast result of the power station in the future (such as 48 hours).

[0081] Part II: Energy Storage Power Station Optimization Strategy Based on Load Forecasting

[0082] The load forecast results are used as input, combined with information such as transformer capacity, energy storage power, energy storage capacity, and electricity price, and the objective function is constructed with the goal of maximizing the operating income of the energy storage power station. Through the pyscipopt module of python, a model is established for optimization and solution to obtain the optimal charging and discharging strategy of the power station. At the same time, the optimization results are sent to the EMS through the cloud, and the energy storage executes the charging and discharging strategy to maximize the benefits.

[0083] The following is combined with Figure 3 Further explanation of this application example

[0084] Step 301: Obtain basic data such as historical load data, holiday information, electricity price information, and historical weather information.

[0085] Step 302: Build a load forecasting model set, including a time series model (prophet), a tree model (xgoost), and a combination model (prophet+xgboost).

[0086] That is to say, a set of load forecasting models is constructed, including a load forecasting model with a time series network structure, a load forecasting model with a tree network structure, and a load forecasting model with a combined network structure including a time series network structure and a tree network structure.

[0087] Step 303: Input historical load data and holiday information into the prophet model, and obtain the parameters corresponding to the model accuracy greater than 85% through model hyperparameter tuning. The accuracy calculation method is MAPE (Mean Absolute Percentage Error):

[0088]

[0089] Where n represents the number of samples, y real represents the measured value, y predict Represents the predicted value.

[0090] The following is an example set of prophet prediction model parameters:

[0091] holidays: legal holidays, #holiday dates;

[0092] changepoint_prior_scale: 0.5, #control the sensitivity of the change point;

[0093] holidays_prior_scale=10, #Controls the sensitivity of holiday effects.

[0094] The holiday information for a period of time in the future is input into the prophet prediction model to obtain the predicted load of the prophet prediction model for the corresponding period of time.

[0095] Step 304: Input historical load data, holiday information, electricity price information, and historical weather information into the xgboost prediction model, and obtain hyperparameters corresponding to a model accuracy greater than 85% through hyperparameter tuning.

[0096] The following is a set of xgboost prediction model hyperparameter examples:

[0097] 'max_depth':6,#The maximum depth of the tree;

[0098] 'eta': 0.05, #learning rate;

[0099] 'objective':'reg:squarederror',#loss function;

[0100] 'eval_metric':'rmse',#Evaluation indicator;

[0101] 'n_estimator':200,#Number of trees;

[0102] 'gamma': 0, #Minimum loss reduction, used to control the complexity of the tree. The larger the value, the more conservative the model.

[0103] 'subsample': 0.8, #The sample ratio used to train each tree to reduce overfitting;

[0104] 'colsample_bytree': 0.8, #The column sample ratio used when building the tree to reduce overfitting.

[0105] The holiday information, electricity price information, and weather information for a period of time in the future are input into the xgboost prediction model to obtain the predicted load of the xgboost prediction model for the corresponding time period.

[0106] Step 305: Input historical load data, holiday information, electricity price information, and historical weather information into the prophet-xgboost prediction model, and obtain the parameters corresponding to the model accuracy greater than 85% through parameter tuning. Input holiday information, electricity price information, and weather information for a period of time in the future into the prophet-xgboost prediction model to obtain the predicted load of the prophet-xgboost prediction model for the corresponding time period.

[0107] The following is an example of a set of prophet-xgboost prediction model hyperparameters:

[0108] holidays: legal holidays, #holiday dates;

[0109] changepoint_prior_scale: 0.5, #control the sensitivity of the change point;

[0110] holidays_prior_scale=10,#Control the sensitivity of holiday effects;

[0111] 'max_depth':6,#The maximum depth of the tree;

[0112] 'eta': 0.05, #learning rate;

[0113] 'objective':'reg:squarederror',#loss function;

[0114] 'eval_metric':'rmse',#Evaluation indicator;

[0115] 'n_estimator':200,#Number of trees;

[0116] 'gamma': 0, #Minimum loss reduction, used to control the complexity of the tree. The larger the value, the more conservative the model.

[0117] 'subsample': 0.8, #The sample ratio used to train each tree to reduce overfitting;

[0118] 'colsample_bytree': 0.8, #The column sample ratio used when building the tree to reduce overfitting.

[0119] Step 306: Calculate the accuracy (MAPE) of each prediction model in real time through measured load data. Count the average prediction accuracy of each prediction model over a period of history (such as 3 days), and select the load prediction result of the prediction model with the highest accuracy for a period of time in the future (such as 48 hours) as the load prediction result of the power station.

[0120] Step 307: Obtain basic information of the power station, including transformer capacity / demand information, energy storage power, energy storage capacity, electricity price information, etc.

[0121] Step 308: Establish an objective function (f) with the goal of maximizing the charging and discharging benefits of the power station.

[0122]

[0123] Among them, P i,charge represents the charging power at the i-th moment, P i,discharge Represents the discharge power at the i-th moment (at the same moment, P i,charge and P i,discharge At least one of them is 0, that is, the energy storage is either charged or discharged at the same time, not charged and discharged at the same time), E i represents the electricity price at the i-th moment, and n is the number of samples.

[0124] Step 309: adding constraints, including energy storage SOC constraints, charge and discharge power constraints, transformer capacity constraints, etc.

[0125] The constraints include:

[0126] SOC constraint: 0 ≤ SOC i ≤1, SOC at any time i i Between 0 and 1

[0127] Charging power constraint: 0≤P i,charge ≤P max , the charging power P at any time i i,charge All between 0 and maximum power P max between

[0128] Discharge power constraint:

[0129] ①0≤P i,discharge ≤P max , discharge power P at any time i i,discharge All between 0 and maximum power O max between

[0130] ②P i,load -P i,discharge ≥0, discharge power P at any time i i,discharge Less than or equal to load power P i,load .

[0131] Transformer capacity constraint: 0≤P i,load +P i,charge ≤P transformer , load power P at any time i,load With charging power P i,charge The sum is less than or equal to the transformer capacity P transformer .

[0132] Step 310: Use the pyscipopt module of Python to build a model, add the constraints in step 309, and use the objective function in step 308 as the optimization target to perform optimization. The optimal operation strategy of the energy storage power station is obtained when the load is predicted in step 306. The following is an example result:

[0133] Model input:

[0134] Maximum energy storage power: p_max = 375kW

[0135] Transformer capacity: H0 = 800 kVA

[0136] Maximum energy storage capacity: SS = 750 kWh

[0137] Electricity price for one day:

[0138] price=[0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0 .2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0. 2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.2541,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.04 05,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405,1.0405 ,1.0405,1.0405,1.0405,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234,0.6234].

[0139] One day's load:

[0140] w_day=[303.78,303.78,303.78,310.18,311.98,307.17,307.17,307.17,308.97,307.17,303.78, 303.78,266.54,298.36,217.75,296.87,246.57,242.28,199.03,199.03,219.67,241.78,247.44, 191.6,148.85,255.37,272.19,342.77,410.06,488.71,410.4,471.72,449.9,513.62,424.42,608.72,446.75,543.19,439.14,537.12,533.97,321.58,332.42,506.28,377.61,506.99,505.14,489 .55,254.67,158.77,176.3,171.78,193.03,197.14,537.07,548.92,616.83,629.03,622.65,607.68,658.61,520.83,415.31,430.93,346.73,454.22,418.62,435.35,435.82,365.96,511.14,344. .02,304.36,297.92,297.92,292.36,297.92,292.36,297.92,292.36,285.11,290.66].

[0141] Model output:

[0142] p_total=[-375.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0, 0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,-301.9194,-37 5.0,-375.0,-375.0,-375.0,-311.29,-375.0,-328.28,0.0,0.0,0.0,0 .0,0.0,0.0,0.0,0.0,0.0,234.8592,332.42,375.0,375.0,375.0,375. 0,375.0,-375.0,-375.0,-375.0,-375.0,-375.0,-262.93,-251.08,195.0,375.0,375.0,375.0,375.0,375.0,375.0,375.0,375.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,-0.0,0.0,0.0,0.0,0.0,0.0].

[0143] According to calculations, the charging and discharging strategy based on the load forecast can help a power station increase its daily revenue by 10% to 20%. Figure 4 shown.

[0144] Step 311: Push the optimal strategy obtained in step 310 to EMS (energy storage management system) through the cloud. EMS controls the energy storage to execute the charging and discharging strategy to maximize the benefits of the energy storage power station. Figure 5 shown.

[0145] The embodiment of the present application obtains the historical data related to charging and discharging of the energy storage power station, including historical load data, holiday information, electricity price information and historical weather information; constructs multiple load forecasting models, and uses the historical data related to charging and discharging to train each load forecasting model to obtain multiple trained load forecasting models, and selects the optimal load forecasting model from the multiple trained load forecasting models; uses the optimal load forecasting model to predict the predicted load results of the energy storage power station in the future preset time period, and uses the predicted load results of the energy storage power station in the future preset time period and the charging and discharging power optimization model constructed by the pyscipopt module of python to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future. The charging and discharging optimization scheme of the energy storage power station provided by the embodiment of the present invention reduces the user's electricity cost, improves the operating income of the energy storage power station, and better serves the operation and development of industrial and commercial energy storage power stations while improving the accuracy and reliability of the power station load forecast.

[0146] Corresponding to the above method embodiment, the present application also provides an embodiment of a charging and discharging optimization device for an energy storage power station, Figure 2 FIG. 1 is a schematic diagram showing the structure of a charging and discharging optimization device for an energy storage power station provided by an embodiment of the present application. Figure 2 As shown, the device comprises:

[0147] The acquisition module 201 is configured to acquire historical data related to charging and discharging of the energy storage power station including historical load data, holiday information, electricity price information and historical weather information;

[0148] The model optimization module 202 is configured to construct multiple load prediction models, and train each load prediction model using the historical data related to charging and discharging to obtain multiple trained load prediction models, and select the optimal load prediction model from the multiple trained load prediction models;

[0149] The charging and discharging optimization module 203 is configured to use the optimal load prediction model to predict the predicted load results of the energy storage power station within a preset time period in the future, and use the predicted load results of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

[0150] The multiple load forecasting models include: a load forecasting model of a time series network structure, a load forecasting model of a tree network structure, and a load forecasting model of a combined network structure including a time series network structure and a tree network structure.

[0151] The above is a schematic scheme of a charge and discharge optimization device of an energy storage power station of this embodiment. It should be noted that the technical scheme of the charge and discharge optimization device of the energy storage power station and the technical scheme of the charge and discharge optimization method of the energy storage power station mentioned above belong to the same concept, and the details not described in detail in the technical scheme of the charge and discharge optimization device of the energy storage power station can be referred to the description of the technical scheme of the charge and discharge optimization method of the energy storage power station mentioned above.

[0152] Figure 6 The block diagram of a computing device 400 according to an embodiment of the present application is shown. The components of the computing device 400 include but are not limited to a memory 410 and a processor 420. The processor 420 is connected to the memory 410 via a bus 430, and the database 450 is used to store data.

[0153] The computing device 400 also includes an access device 440 that enables the computing device 400 to communicate via one or more networks 460. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 440 may include one or more of any type of network interface (e.g., a network interface card (NIC)) wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0154] In one embodiment of the present application, the above components of the computing device 400 and Figure 6 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 6 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of the present application. Those skilled in the art may add or replace other components as needed.

[0155] Computing device 400 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. Computing device 400 may also be a mobile or stationary server.

[0156] The processor 420 is used to execute the following computer executable instructions, which, when executed by the processor, implement the steps of the above-mentioned method for optimizing charging and discharging of the energy storage power station.

[0157] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned energy storage power station charging and discharging optimization method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above-mentioned energy storage power station charging and discharging optimization method.

[0158] An embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the above-mentioned method for optimizing charging and discharging of an energy storage power station.

[0159] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned energy storage power station charging and discharging optimization method belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the above-mentioned energy storage power station charging and discharging optimization method.

[0160] An embodiment of the present application further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned method for optimizing charging and discharging of an energy storage power station.

[0161] The above is a schematic scheme of a computer program of this embodiment. It should be noted that the technical scheme of the computer program and the technical scheme of the above-mentioned energy storage power station charging and discharging optimization method belong to the same concept, and the details not described in detail in the technical scheme of the computer program can be referred to the description of the technical scheme of the above-mentioned energy storage power station charging and discharging optimization method.

[0162] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0163] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0164] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited by the described action sequence, because according to the embodiments of the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of the present application.

[0165] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0166] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The optional embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the embodiments of the present application. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of the present application, so that those skilled in the art can understand and use the present application well. The present application is only limited by the claims and their full scope and equivalents.

Claims

1. A charging and discharging optimization method for an energy storage power station, characterized in that: include: Obtain historical data related to charging and discharging of energy storage power stations, including historical load data, holiday information, electricity price information, and historical weather information; Constructing multiple load prediction models, and using the historical data related to charging and discharging to train each load prediction model to obtain multiple trained load prediction models, and selecting the optimal load prediction model from the multiple trained load prediction models; The optimal load forecasting model is used to predict the predicted load results of the energy storage power station within a preset time period in the future, and the predicted load results of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python are used to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

2. The charge and discharge optimization method according to claim 1, characterized in that: The multiple load forecasting models include: a load forecasting model of a time series network structure, a load forecasting model of a tree network structure, and a load forecasting model of a combined network structure including a time series network structure and a tree network structure.

3. The charge and discharge optimization method according to claim 2, characterized in that: The method of training each load prediction model by using the historical data related to charging and discharging to obtain multiple trained load prediction models includes: Using the historical load data and holiday information in the historical data related to charging and discharging to train the load forecasting model of the time series network structure, to obtain a trained load forecasting model of the time series network structure; Using the historical load data, holiday information, electricity price information and historical weather information in the historical data related to charging and discharging to train the load forecasting model of the tree network structure, to obtain a trained load forecasting model of the tree network structure; The load forecasting model of the combined network structure is trained using the historical load data, holiday information, electricity price information and historical weather information in the historical data related to charging and discharging, so as to obtain a trained load forecasting model of the combined network structure.

4. The charge and discharge optimization method according to claim 3, characterized in that: The step of selecting the optimal load forecasting model from the trained multiple load forecasting models includes: Calculate in real time a first real-time prediction accuracy of the load prediction model of the time series network structure, and calculate a first average prediction accuracy of the load prediction model of the time series network structure based on the first real-time prediction accuracy; Calculating in real time a second real-time prediction accuracy of the load prediction model of the tree network structure, and calculating a second average prediction accuracy of the load prediction model of the tree network structure according to the second real-time prediction accuracy; Calculate in real time a third real-time prediction accuracy of the load prediction model of the combined network structure, and calculate a third average prediction accuracy of the load prediction model of the combined network structure based on the third real-time prediction accuracy; According to the first average prediction accuracy, the second average prediction accuracy and the third average prediction accuracy, a load prediction model with the largest average prediction accuracy is selected as the optimal load prediction model.

5. The charge and discharge optimization method according to claim 1, characterized in that: The optimal charging and discharging power of the energy storage power station at each moment in the future is obtained by using the predicted load result of the energy storage power station in the future preset time period and the charging and discharging power optimization model constructed by the pyscipopt module of python, including: Obtaining basic information of the energy storage power station including transformer capacity, energy storage power, energy storage capacity and electricity price information; Taking maximizing the charging and discharging benefits of the energy storage power station as the goal, constructing an objective function; Construct constraints including energy storage SOC constraints, charging power constraints, discharging power constraints and transformer capacity constraints; Constructing a charge-discharge power optimization model through the pyscipopt module of python, and adding the objective function and the constraint condition to the charge-discharge power optimization model to obtain a target charge-discharge power optimization model; By inputting the basic information of the energy storage power station including transformer capacity, energy storage power, energy storage capacity and electricity price information and the predicted load results of the energy storage power station in a future preset time period into the target charging and discharging power optimization model, the optimal charging and discharging power of the energy storage power station at each moment in the future is obtained.

6. The charge and discharge optimization method according to claim 5, characterized in that: The constraints include: The energy storage SOC constraint is that the SOC at any time is between 0 and 1; The charging power constraint is that the charging power at any time is between 0 and the maximum charging power; The discharge power constraint is that the discharge power at any time is between 0 and the maximum discharge power and the discharge power at any time is not greater than the load power; The transformer capacity constraint is that the transformer capacity at any time is greater than or equal to the sum of the load power and the charging power.

7. A charging and discharging optimization device for an energy storage power station, characterized in that: include: An acquisition module is configured to acquire historical data related to charging and discharging of the energy storage power station including historical load data, holiday information, electricity price information and historical weather information; A model optimization module is configured to construct multiple load prediction models, and train each load prediction model using the historical data related to charging and discharging to obtain multiple trained load prediction models, and select the optimal load prediction model from the multiple trained load prediction models; The charging and discharging optimization module is configured to use the optimal load prediction model to predict the predicted load results of the energy storage power station within a preset time period in the future, and use the predicted load results of the energy storage power station within the preset time period in the future and the charging and discharging power optimization model constructed by the pyscipopt module of python to obtain the optimal charging and discharging power of the energy storage power station at each moment in the future.

8. The charge and discharge optimization device according to claim 7, characterized in that: The multiple load forecasting models include: a load forecasting model of a time series network structure, a load forecasting model of a tree network structure, and a load forecasting model of a combined network structure including a time series network structure and a tree network structure.

9. A computing device comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the charging and discharging optimization method of the energy storage power station described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for optimizing charging and discharging of an energy storage power station as claimed in any one of claims 1 to 6.

Images