Energy management method and device for optical storage system

Abstract

The invention provides an energy management method and device for an optical storage system. In the energy management method, factors such as time-of-use electricity price, battery durability and the like are considered in a constructed energy management optimization model, a power prediction value of photovoltaic power generation is used as input, an output plan of an energy storage battery in the optical storage system is obtained through prediction, and then the energy storage battery is scheduled according to the output plan of the energy storage battery. According to the method, when the electricity selling price of the power grid is high and the photovoltaic power generation is insufficient, the load electricity in the optical storage system can be supplemented through the energy storage battery, so that the electricity purchasing cost is saved; when the photovoltaic power generation is sufficient and the electricity sold by the power grid is at a high price, the electricity can be sold to the power grid to obtain high benefits; when the electricity selling price of the power grid is low, electricity can be purchased from the power grid to supplement the load electricity in the optical storage system; therefore, the operation efficiency of the optical storage system, the utilization rate of renewable energy sources and the economical efficiency of the system are improved, and the applicability is better.

Description

Technical Field

[0001] The present application relates to the field of new energy management technology, and in particular to an energy management method and device for a photovoltaic storage system. Background Art

[0002] With the development of clean energy and energy storage technologies, the diversified development of energy has led to the direct coupling of multiple devices in the energy system, which has restricted the flexible operation of the system and caused the mismatch between the supply and demand of energy in the system. The photovoltaic storage system can effectively utilize photovoltaic power generation while smoothing the photovoltaic power output through energy storage batteries, which can improve the reliability of the system power supply and solve the problem of the constraints of multiple device coupling on the system.

[0003] However, in order to effectively exert the function of energy storage batteries and thus improve the operating efficiency of the photovoltaic storage system, accurate scheduling of energy storage batteries is crucial. Therefore, how to accurately schedule the energy storage batteries in the photovoltaic storage system to improve the operating efficiency of the photovoltaic storage system has become a technical problem to be solved in this field. Summary of the invention

[0004] In order to solve the problem of how to accurately schedule energy storage batteries in a photovoltaic storage system to improve the operating efficiency of the photovoltaic storage system, the present application provides an energy management method and device for a photovoltaic storage system.

[0005] In a first aspect, an embodiment of the present application provides an energy management method for a photovoltaic storage system, the method comprising: determining real number variables and binary variables for constructing an energy management optimization model for the photovoltaic storage system; the real number variables include the output of the energy storage battery, the state of charge of the energy storage battery, and the power grid output; the binary variables include the operating state of the energy storage battery, and the operating state of the energy storage battery is used to characterize that the energy storage battery is in a charging state / discharging state; based on the real number variables and the binary variables, an objective function is constructed with the goal of minimizing the total system cost and minimizing the life attenuation of the energy storage battery; the total system cost includes the power grid purchase and sale cost, the photovoltaic power generation maintenance cost, and the energy storage power Pool attenuation loss cost; determine the optimization constraints based on the real variables and binary variables; the optimization constraints include the power balance constraints of the photovoltaic storage system, the output constraints of the energy storage battery, and the output constraints of the power grid; based on the objective function and the optimization constraints, construct a non-convex mixed integer quadratic constrained nonlinear programming problem model as the energy management optimization model of the photovoltaic storage system; obtain the power prediction value of photovoltaic power generation in the photovoltaic storage system; based on the power prediction value, optimize and solve the energy management optimization model of the photovoltaic storage system to obtain the output plan of the energy storage battery in the photovoltaic storage system; and operate and regulate the energy storage battery based on the output plan.

[0006] Second aspect, an embodiment of the present application further provides an energy management system for a photovoltaic-storage system. The system includes: an energy management server; the energy management server is configured to: determine real variables and binary variables for constructing an energy management optimization model of the photovoltaic-storage system; the real variables include the output of the energy storage battery, the state of charge of the energy storage battery, and the output of the power grid; the binary variables include the operating state of the energy storage battery, and the operating state of the energy storage battery is used to represent that the energy storage battery is in a charging state / discharging state; based on the real variables and binary variables, construct an objective function with the minimization of the total system cost and the minimization of the energy storage battery life attenuation as the objectives; the total system cost includes the power grid purchase and sale cost, the photovoltaic power generation maintenance cost, and the energy storage battery attenuation loss cost; determine optimization constraint conditions based on the real variables and binary variables; the optimization constraint conditions include the power balance constraint condition of the photovoltaic-storage system, the output constraint condition of the energy storage battery, and the output constraint condition of the power grid; based on the objective function and the optimization constraint conditions, construct a non-convex mixed-integer quadratic constraint non-linear programming problem model as the energy management optimization model of the photovoltaic-storage system; obtain the power prediction value of the photovoltaic power generation in the photovoltaic-storage system; based on the power prediction value, optimize and solve the energy management optimization model of the photovoltaic-storage system to obtain the output plan of the energy storage battery in the photovoltaic-storage system; perform operation control on the energy storage battery based on the output plan.

[0007] Third aspect, an embodiment of the present application further provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the method of the first aspect is implemented.

[0008] Fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by the processor, the method of the first aspect is implemented.

[0009] An embodiment of the present application provides an energy management method and device for a photovoltaic-storage system. In the method provided by the present application, the constructed energy management optimization model takes into account factors such as time-of-use electricity price and battery durability, uses the power prediction value of the photovoltaic power generation as input, predicts the output plan of the energy storage battery in the photovoltaic-storage system, and then schedules the energy storage battery according to the output plan of the energy storage battery. Through this method, when the power grid sells electricity at a high price and the photovoltaic power generation is insufficient, the energy storage battery can be used to supplement the load electricity in the photovoltaic-storage system to save the electricity purchase cost; when the photovoltaic power generation is sufficient and the power grid sells electricity at a high price, electricity can be sold to the power grid to obtain high profits; when the power grid sells electricity at a low price, electricity can be purchased from the power grid to supplement the load electricity in the photovoltaic-storage system; thereby improving the operating efficiency of the photovoltaic-storage system, the utilization rate of renewable energy, and the economy of the system, and having better applicability. Description of the Drawings

[0010] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0011] Figure 1 It is a schematic flowchart of an energy management method for a photovoltaic energy storage system provided by an embodiment of the present application.

[0012] Figure 2 It is a schematic flowchart of a model training method provided by an embodiment of the present application.

[0013] Figure 3 It is a schematic diagram of an application scenario provided by an embodiment of the present application.

[0014] Figure 4 It is another schematic diagram of an application scenario provided by an embodiment of the present application.

[0015] Figure 5 It is a block diagram of the structure of an energy management system for a photovoltaic energy storage system provided by an embodiment of the present application.

[0016] Figure 6 It is a block diagram of the structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0017] The following further details the present application through the accompanying drawings and embodiments. Through these descriptions, the features and advantages of the present application will become clearer and more distinct.

[0018] The special term "exemplary" here means "serving as an example, an embodiment, or illustrative". Any embodiment described as "exemplary" here does not have to be interpreted as superior to or better than other embodiments. Although various aspects of the embodiments are shown in the accompanying drawings, unless otherwise specified, the accompanying drawings do not have to be drawn to scale.

[0019] In addition, the technical features involved in different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0020] The following describes the embodiments of the energy management method for a photovoltaic energy storage system provided by the present application in conjunction with the accompanying drawings.

[0021] See Figure 1 , Figure 1The flowchart shows a method for energy management in a photovoltaic energy storage system provided by an embodiment of the present application. This method can be applied to terminal devices, servers, etc. Hereinafter, taking a server as an example, the method for energy management in a photovoltaic energy storage system provided by an embodiment of the present application will be described. As Figure 1 shown, the method may include the following steps:

[0022] Step S101: Determine real variables and binary variables for constructing an energy management optimization model of the photovoltaic energy storage system.

[0023] Among them, the real variables may include the output of the energy storage battery, the state of charge (SOC) of the energy storage battery, and the output of the power grid. The output of the energy storage battery may also be referred to as the charge and discharge power of the energy storage battery. The output of the power grid may also be referred to as the grid interaction power, and the present application does not limit this.

[0024] The binary variables may include the operating state of the energy storage battery, and the operating state of the energy storage battery is used to represent that the energy storage battery is in a charging state / discharging state. The binary variable may also be referred to as a 0-1 variable. Among them, when the energy storage battery discharges, the binary variable of the operating state of the energy storage battery is equal to 1, and when the energy storage battery charges, the binary variable of the operating state of the energy storage battery is equal to 0.

[0025] It should be understood that the real variables and binary variables for constructing the energy management optimization model of the photovoltaic energy storage system may both include other parameter variables, which will not be listed one by one here.

[0026] Step S102: Based on the real variables and binary variables, construct an objective function with the goal of minimizing the total system cost and minimizing the life attenuation of the energy storage battery.

[0027] Among them, the total system cost may include the cost of buying and selling electricity from the power grid, the cost of photovoltaic power generation maintenance, and the cost of energy storage battery attenuation loss.

[0028] The inventors found in actual applications and research that in the case of long-term high-power and high-depth discharge, the service life of the energy storage battery will accelerate attenuation, and among them, the depth of discharge (DOD) is an important factor affecting the service life of the energy storage battery.

[0029] Based on this, in an optional embodiment of the present application, before constructing the objective function, the server will first determine the service life of the energy storage battery based on the depth of discharge of the energy storage battery, and then determine the loss cost of the energy storage battery, so as to determine the total system cost subsequently.

[0030] Optionally, the service life of the energy storage battery after the i-th discharge can be determined according to the following formula (1).

[0031]

[0032] Among them, in formula (1), N life,i represents the service life of the energy storage battery after the i-th discharge; D od,i represents the depth of discharge of the energy storage battery during the i-th discharge; α1, α2, α3, α4, α5 represent fitting parameters; i is a positive integer.

[0033] Optionally, after determining the service life of the energy storage battery after the i-th discharge, the server can determine the loss cost of the i-th discharge of the energy storage battery based on the service life of the energy storage battery after the i-th discharge according to the following formula (2).

[0034]

[0035] Among them, in formula (2), γ%(i) represents the loss cost of the i-th discharge of the energy storage battery.

[0036] After that, considering factors such as the life attenuation of the energy storage battery and the time-of-use electricity price, when constructing the energy management optimization model of the photovoltaic energy storage system, the economy of the system and the durability of the energy storage battery are taken into account, and an objective function with the minimization of the total system cost (or the total operating cost of the system) and the minimization of the life attenuation of the energy storage battery as the goal is constructed.

[0037] Optionally, the objective function may include the following relational expressions (3)-(6).

[0038]

[0039] C pv = ξ pv ×E pv (6)

[0040] Among them, in formulas (3)-(6), C cost represents the total system cost; C grid represents the grid power purchase and sale cost; C pv represents the photovoltaic power generation maintenance cost; C bat represents the energy storage battery attenuation loss cost; t represents the t-th sampling moment; T represents the total number of samplings; both t and T are positive integers; S buy (t) represents the electricity price at which the photovoltaic energy storage system purchases electricity from the grid at the t-th sampling moment; S sell (t) represents the electricity price at which the photovoltaic energy storage system sells electricity to the grid at the t-th sampling moment; P grid represents the grid power output; b represents the unit cost of the energy storage battery; E bat represents the capacity of the energy storage battery; γ%(t) represents the loss cost of the energy storage battery discharge at the t-th sampling moment; ξ pv represents the photovoltaic power generation maintenance coefficient; E pvRepresents the installed photovoltaic capacity.

[0041] Step S103: Determine the optimization constraint conditions based on the real number variables and binary variables.

[0042] Among them, the optimization constraint conditions may include the power balance constraint condition of the photovoltaic energy storage system, the output constraint condition of the energy storage battery, and the grid output constraint condition (or called the grid interaction power constraint condition).

[0043] Optionally, the power balance constraint condition of the photovoltaic energy storage system may include the following relational expression (7).

[0044] P pv +P bat +P grid =P L (7)

[0045] Among them, in the relational expression (7), P pv represents the predicted power value of photovoltaic power generation; P bat represents the output of the energy storage battery; P grid represents the grid output; P L represents the load power.

[0046] Optionally, the output constraint condition of the energy storage battery may include the following relational expression (8).

[0047]

[0048] Among them, in the relational expression (8), λ represents the operating state of the energy storage battery, which is a binary variable (or called a 0-1 variable). When the energy storage battery discharges, λ is equal to 1, and when the energy storage battery charges, λ is equal to 0; P bat,min represents the lower limit of the output of the energy storage battery; P bat,max represents the upper limit of the output of the energy storage battery; P bat (:,k) represents the output of the energy storage battery at the kth sampling moment, and k is a positive integer.

[0049] Optionally, the grid output constraint condition may include the following relational expression (9).

[0050] P grid,min ≤P grid ≤P grid,max (9)

[0051] Among them, in the relational expression (9), P grid,min represents the lower limit of the grid output, and P grid,max represents the upper limit of the grid output.

[0052] Step S104: Based on the objective function and the optimization constraints, construct a non-convex mixed-integer quadratically constrained nonlinear programming problem model as the energy management optimization model of the photovoltaic energy storage system.

[0053] Among them, the non-convex mixed-integer quadratically constrained nonlinear programming problem model can be abbreviated as a non-convex MIQCP (mixed-integer quadratically constrained programming) model. After the server constructs the objective function and the optimization constraints, it can construct a non-convex MIQCP model based on the objective function and the optimization constraints, and use the constructed non-convex MIQCP model as the energy management optimization model of the photovoltaic energy storage system.

[0054] Step S105: Obtain the power prediction value of photovoltaic power generation in the photovoltaic energy storage system.

[0055] After the server constructs the energy management optimization model of the photovoltaic energy storage system, it can use the power prediction value of photovoltaic power generation in the photovoltaic energy storage system as the input of this energy management optimization model to predict the output of the energy storage battery in the photovoltaic energy storage system.

[0056] In an optional embodiment of the present application, obtaining the power prediction value of photovoltaic power generation in the photovoltaic energy storage system can be implemented in the following manner:

[0057] The first step: Use the improved non-dominated sorting genetic algorithm (or called NSGA-II multi-objective genetic algorithm) based on the tent map (or called Tent chaotic map) to optimize the hyperparameters of the learning rate, the number of neurons, and the regularization coefficient of the pre-established convolutional neural network - bidirectional long short-term memory network - attention mechanism (CNN-BiLSTM-Attention) hybrid deep learning model to obtain an optimized model.

[0058] The second step: Train the optimized model to obtain a trained power prediction model.

[0059] See Figure 2 , Figure 2 which is a schematic flow chart of a model training method provided by an embodiment of the present application. The optimized model can be trained according to the method shown in Figure 2 to obtain a trained power prediction model. As shown in Figure 2 , this model training method may include the following steps:

[0060] Step S201: Obtain a training data set.

[0061] Among them, the training data set includes the first historical meteorological data sequence of the first meteorological parameter, the first historical photovoltaic power generation data sequence, and the first time stamp data sequence. The first meteorological parameter is a meteorological parameter whose correlation with the photovoltaic power generation is greater than the correlation threshold.

[0062] Optionally, obtaining the training data set can be implemented in the following manner:

[0063] Step 1: Collect historical data according to a preset collection frequency and construct a historical data set.

[0064] Among them, the historical data may include the historical meteorological data of the second meteorological parameter, the historical photovoltaic power generation data, and the time stamp data. The second meteorological parameter refers to a meteorological parameter that is related to the photovoltaic power generation and has an impact on the photovoltaic power generation. The second meteorological parameter includes the first meteorological parameter.

[0065] Optionally, the second meteorological parameter may include total solar irradiance (TSI), direct normal irradiance (DNI), global horizontal irradiance (GHI), wind speed, wind direction, temperature, humidity, air pressure, etc.

[0066] Optionally, the historical meteorological data can be collected by a multi-element meteorological instrument.

[0067] Optionally, the preset collection frequency can be set according to the requirements of the actual application scenario. For example, the collection frequency can be set to 15 minutes (min). That is, collect historical data at a frequency of once every 15 minutes.

[0068] Then, the collected historical data can be constructed into a historical data set in the order of the corresponding collection time. The historical data set may include the second historical meteorological data sequence of the second meteorological parameter, the second historical photovoltaic power generation data sequence, and the second time stamp data sequence.

[0069] Step 2: Preprocess the historical data included in the historical data set to obtain a preprocessed data set.

[0070] Among them, the preprocessing may include data cleaning, filling, and other processing.

[0071] Step 3: Use the Spearman rank correlation coefficient method to determine the correlation between the second meteorological parameter and the photovoltaic power generation, and screen out the first meteorological parameter whose correlation with the photovoltaic power generation is greater than the correlation threshold from the second meteorological parameter.

[0072] The Spearman rank correlation coefficient method (abbreviated as the Spearman method) can be used to determine the correlation between each second meteorological parameter and the photovoltaic power generation, so as to select the first meteorological parameter with a higher correlation with the photovoltaic power generation, reduce the interference of non-critical characteristic parameters, and make the subsequent power prediction more accurate.

[0073] Step 4: Based on the first meteorological parameter and the preprocessed data set, construct the training data set.

[0074] After determining the first meteorological parameter, the historical meteorological data sequence corresponding to the first meteorological parameter can be selected from the preprocessed data set and denoted as the first historical meteorological data sequence. And the time stamp data sequence in the preprocessed data set can be denoted as the first time stamp data sequence, and the historical photovoltaic power generation data sequence in the preprocessed data set can be denoted as the first historical photovoltaic power generation data sequence. Then, the training data set is constructed by using the first historical meteorological data sequence, the first historical photovoltaic power generation data sequence and the first time stamp data sequence.

[0075] Step S202: Use the Gaussian mixture model clustering method to perform clustering analysis on the training data set, and divide the training data set into a sunny day data set, a cloudy day data set and a rainy day data set based on the clustering analysis results.

[0076] After performing clustering analysis on the training data set by using the Gaussian mixture model (GMM) clustering method, the training data set can be divided into a sunny day data set, a cloudy day data set and a rainy day data set based on the clustering analysis results.

[0077] Step S203: Generate sunny day samples, cloudy day samples and rainy day samples based on the sunny day data set, the cloudy day data set and the rainy day data set respectively.

[0078] Based on the collection time corresponding to the data in the sunny day data set, the cloudy day data set and the rainy day data set, the data corresponding to each date can be screened out, and then based on all the data corresponding to each date, in the order of the collection time, the daily samples corresponding to the respective dates can be constructed.

[0079] Step S204: Use the composite mode symmetry empirical mode decomposition with adaptive noise algorithm to decompose each daily sample into components with different frequencies.

[0080] After obtaining sunny-day samples, cloudy-day samples, and rainy-day samples, the cross-correlation enhanced multi-dimensional adaptive network (CCEMDAN) algorithm can be used to decompose each daily sample into components of different frequencies. That is, each sunny-day sample is decomposed into components of different frequencies, each cloudy-day sample is decomposed into components of different frequencies, and each rainy-day sample is decomposed into components of different frequencies.

[0081] Step S205: In an iterative manner, input each component of each daily sample into the optimization model in sequence to train the optimization model and obtain a trained power prediction model.

[0082] Among them, during any iteration, after inputting each component of the corresponding daily sample into the optimization model, the predicted values of each component are obtained, and the power prediction value of the corresponding photovoltaic power generation for the daily sample is obtained by superimposing the predicted values of each component.

[0083] For example, after inputting each component of a certain sunny-day sample into the optimization model, the predicted values corresponding to each component can be obtained using the optimization model. Then, after weighted superposition of the predicted values corresponding to each component, the weighted superposition result can be used as the power prediction value of the photovoltaic power generation corresponding to this sunny-day sample.

[0084] Use different daily samples to train the optimization model in sequence until the model converges, and a trained power prediction model can be obtained.

[0085] Third step: Use the power prediction model to predict the power prediction value of the photovoltaic power generation in the photovoltaic energy storage system.

[0086] After obtaining the trained power prediction model, the power prediction value of the photovoltaic power generation in the photovoltaic energy storage system can be predicted using this power prediction model.

[0087] As described above, during the process of training the power prediction model, the acquisition frequency can be set according to the requirements of the actual application scenario. Correspondingly, when establishing the CNN-BiLSTM-Attention hybrid deep learning model, the sampling frequency of the model prediction can be set according to the prediction requirements. Optionally, in order to improve the prediction accuracy of the model, the sampling frequency of the model prediction can be set to the aforementioned acquisition frequency. For example, the sampling frequency of the model prediction can be set to 15 minutes. In this case, if we want to predict the photovoltaic power generation for the next 24 hours, 96 prediction values can be obtained. That is, if the next 24 hours is one day in the future, a total of 96 prediction values can be obtained from 0:00:00 to 24:00:00 in the future day, and the interval between two adjacent prediction values is 15 minutes.

[0088] That is to say, optionally, after obtaining the trained power prediction model, the power prediction values of the photovoltaic power generation of the energy storage system in the next 24 hours can be predicted using this power prediction model.

[0089] For example, assume that the sunlight is concentrated from 6:00 to 19:00, and the time resolution (or sampling rate) is 15 minutes (min), and predict the photovoltaic power generation for the next 24 hours. The prediction results are as Figure 3 shown. The root mean square error RMSE of the prediction results of the photovoltaic power generation is 0.421 megawatts (MW), the mean absolute error is 0.26 megawatts, and the goodness of fit R 2 is 0.934. The prediction results fit the actual output power curve of the photovoltaic power generation, and the prediction accuracy is relatively high.

[0090] Step S106: Based on the power prediction values, optimize and solve the energy management optimization model of the energy storage system to obtain the output plan of the energy storage battery in the energy storage system.

[0091] Optionally, when optimizing and solving the energy management optimization model of the energy storage system, the non-linear model optimization problem can be converted into a linear model optimization problem by means of piecewise linear programming. Then, use the Yalmip toolbox to call the Gurobi solver for solution to obtain the output plan of the energy storage battery in the energy storage system. Optionally, the output plan of the energy storage battery in the energy storage system can be the output plan of the energy storage battery in the next 24 hours. Optionally, the time resolution (or sampling frequency) in the output plan for the next 24 hours can be 15 minutes.

[0092] Step S107: Perform operation regulation on the energy storage battery based on the output plan.

[0093] After obtaining the output plan of the energy storage battery, the operation of the energy storage battery can be regulated or scheduled based on the output plan of the energy storage battery, such as the output plan of the energy storage battery in the next 24 hours, so as to smooth the output of the photovoltaic power generation and make the photovoltaic-storage system more stable.

[0094] For example, as Figure 4 shown, taking the prediction result in Figure 3 as the input of the energy management optimization model of the photovoltaic-storage system, energy management optimization is carried out. During the first peak electricity price period, the photovoltaic power generation meets the load demand. From 10:00 to 14:00 is the high electricity price period, and the remaining photovoltaic power generation is sold to the power grid. Moreover, according to the method of this application, during the period when the photovoltaic power is insufficient, the energy storage battery can be actively discharged to save the electricity purchase cost. During the period from 15:00 to 18:00, the photovoltaic power generation is low and it is a low electricity price period. The remaining power is supplemented by the power grid, and the energy storage battery absorbs all the photovoltaic power generation. During the evening peak electricity price period, the electric power stored by the energy storage battery during the period from 15:00 to 18:00 effectively meets the load demand and reduces the electricity purchase from the power grid. According to the method of this application, the electricity purchase and sale cost is more appropriate. The charge and discharge scheduling of the energy storage battery makes the attenuation loss cost of the energy storage battery lower, improves the utilization rate of renewable energy, system economy and durability of the energy storage battery, and has good applicability.

[0095] The energy management method for the photovoltaic-storage system provided by the embodiment of this application constructs an energy management optimization model considering factors such as time-of-use electricity price and battery durability, takes the power prediction value of the photovoltaic power generation as the input, predicts the output plan of the energy storage battery in the photovoltaic-storage system, and then schedules the energy storage battery according to the output plan of the energy storage battery. By scheduling the energy storage battery through the method provided by this application, when the power grid sells electricity at a high price and the photovoltaic power generation is insufficient, the load electricity in the photovoltaic-storage system can be supplemented by the energy storage battery to save the electricity purchase cost; when the photovoltaic power generation is sufficient and the power grid sells electricity at a high price, electricity can be sold to the power grid to obtain high benefits; when the power grid sells electricity at a low price, electricity can be purchased from the power grid to supplement the load electricity in the photovoltaic-storage system; thereby improving the operation efficiency of the photovoltaic-storage system, the utilization rate of renewable energy and the economy of the system, and having better applicability.

[0096] In addition, when obtaining the power prediction value of photovoltaic power generation, the NSGA-II algorithm improved based on Tent chaotic mapping is used to optimize the hyperparameters of the learning rate, the number of neurons, and the regularization coefficient of the pre-established GNN-BiLSTM-Attention hybrid deep learning model, obtaining an optimized model. After training the optimized model, a well-trained power prediction model is obtained, improving the prediction accuracy of photovoltaic power generation. Moreover, during the process of training the optimized model, the Spearman method is used for feature extraction and data dimensionality reduction, and the GMM method is used to perform clustering analysis on the samples, thereby dividing the samples into daily samples of three different weather types: sunny, cloudy, and rainy. Then, each daily sample is decomposed into components with different frequencies through the CCEMDAN algorithm. Subsequently, each component of each sample is respectively input into the power prediction model for prediction to obtain the prediction values of each component. After that, the prediction values of each component are superimposed to obtain the power prediction value of the corresponding daily sample of photovoltaic power generation, thus further improving the accuracy and stability of photovoltaic power generation prediction, providing stable support for energy management, and further improving the operation stability of the photovoltaic energy storage system.

[0097] It can be understood that the above embodiments are only examples. During actual implementation, the above embodiments can be modified. Those skilled in the art can understand that the modification methods that do not require creative labor for the above embodiments all fall within the protection scope of this application and will not be elaborated in the embodiments.

[0098] Based on the same inventive concept, the embodiments of this application also provide an energy management system for a photovoltaic energy storage system. Since the principle of the problem solved by the energy management system for a photovoltaic energy storage system is similar to that of the aforementioned energy management method for a photovoltaic energy storage system, the implementation of the energy management system for a photovoltaic energy storage system can refer to the implementation of the aforementioned energy management method for a photovoltaic energy storage system, and the repeated parts will not be elaborated.

[0099] See Figure 5 , Figure 5 which is a structural block diagram of an energy management system for a photovoltaic energy storage system provided by the embodiments of this application. As Figure 5 shown, the energy management system 500 for a photovoltaic energy storage system may include: an energy management server 501 and a coordination controller 502. It should be understood that the energy management system 500 for a photovoltaic energy storage system may also include other devices, such as a digital twin module, a station-side data acquisition device, a monitoring and control system, a switch, a communication gateway, an uninterruptible power supply, etc., which will not be listed one by one here. Among them, the energy management server 501 can be used for:

[0100] Determine the real variables and binary variables for constructing the energy management optimization model of the photovoltaic energy storage system; the real variables include the output of the energy storage battery, the state of charge of the energy storage battery, and the output of the power grid; the binary variable includes the operating state of the energy storage battery, and the operating state of the energy storage battery is used to characterize that the energy storage battery is in the charging state / discharging state;

[0101] Based on the real variables and binary variables, construct an objective function with the minimization of the total system cost and the minimization of the life attenuation of the energy storage battery as the objectives; the total system cost includes the power purchase and sale cost of the power grid, the maintenance cost of photovoltaic power generation, and the attenuation loss cost of the energy storage battery;

[0102] Determine the optimization constraint conditions based on the real variables and binary variables; the optimization constraint conditions include the power balance constraint condition of the photovoltaic energy storage system, the output constraint condition of the energy storage battery, and the output constraint condition of the power grid;

[0103] Based on the objective function and the optimization constraint conditions, construct a non-convex mixed-integer quadratic constraint non-linear programming problem model as the energy management optimization model of the photovoltaic energy storage system;

[0104] Obtain the power prediction value of the photovoltaic power generation in the photovoltaic energy storage system;

[0105] Based on the power prediction value, optimize and solve the energy management optimization model of the photovoltaic energy storage system to obtain the output plan of the energy storage battery in the photovoltaic energy storage system;

[0106] Perform operation regulation on the energy storage battery based on the output plan.

[0107] In a possible implementation manner, the energy management server 501 is used to perform operation regulation on the energy storage battery based on the output plan, including: the energy management server 501 sends the output plan to the coordination controller 502, and performs operation regulation on the energy storage battery through the coordination controller 502.

[0108] In a possible implementation manner, before constructing the objective function, the energy management server 501 is further used to: determine the service life of the energy storage battery after the i-th discharge according to the following formula; where N life,i represents the service life of the energy storage battery after the i-th discharge; D od,i represents the depth of discharge of the energy storage battery in the i-th discharge; α1, α2, α3, α4, α5 represent fitting parameters; i is a positive integer; based on the service life of the energy storage battery after the i-th discharge, determine the loss cost of the i-th discharge of the energy storage battery according to the following formula; where γ%(i) represents the loss cost of the i-th discharge of the energy storage battery.

[0109] In a possible implementation, the objective function includes the following relational expression:

[0110]

[0111] C pv = ξ pv × E pv ;

[0112] where C cost represents the total cost of the system; C grid represents the cost of power purchase and sale of the power grid; C pv represents the maintenance cost of photovoltaic power generation; C bat represents the attenuation loss cost of the energy storage battery; t represents the t-th sampling moment; T represents the total number of samplings; both t and T are positive integers; S buy (t) represents the electricity price for the energy storage and photovoltaic system to purchase electricity from the power grid at the t-th sampling moment; S sell (t) represents the electricity price for the energy storage and photovoltaic system to sell electricity to the power grid at the t-th sampling moment; P grid represents the power output of the power grid; b represents the unit cost of the energy storage battery; E bat represents the capacity of the energy storage battery; γ%(t) represents the loss cost of the energy storage battery discharging at the t-th sampling moment; ξ pv represents the photovoltaic power generation maintenance coefficient; E pv represents the photovoltaic installed capacity.

[0113] In a possible implementation, the power balance constraint condition of the energy storage and photovoltaic system includes the following relational expression: P pv + P bat + P grid = P L ; where P pv represents the predicted value of the photovoltaic power generation; P bat represents the output of the energy storage battery; P grid represents the power output of the power grid; P L represents the load power.

[0114] In a possible implementation, the output constraint condition of the energy storage battery includes the following relational expressions:

[0115]

[0116] where λ represents the operating state of the energy storage battery. When the energy storage battery discharges, λ is equal to 1. When the energy storage battery charges, λ is equal to 0; P bat,min represents the lower limit of the output of the energy storage battery; P bat,max represents the upper limit of the output of the energy storage battery; P bat (:, k) represents the output of the energy storage battery at the k-th sampling moment, and k is a positive integer.

[0117] In a possible implementation, the grid output constraint condition includes the following relational expression: P grid,min ≤P grid ≤P grid,max ; where P grid,min represents the lower limit of the grid output, and P grid,max represents the upper limit of the grid output.

[0118] In a possible implementation, the energy management server 501 is used to obtain the power prediction value of photovoltaic power generation in the photovoltaic-storage system, including: the energy management server 501 is used to: optimize the hyperparameters of the learning rate, the number of neurons, and the regularization coefficient of a pre-established convolutional neural network-bi-directional long short-term memory network-attention mechanism hybrid deep learning model by using the improved second-generation non-dominated sorting genetic algorithm based on the tent map to obtain an optimized model; train the optimized model to obtain a trained power prediction model; and use the power prediction model to predict the power prediction value of photovoltaic power generation in the photovoltaic-storage system.

[0119] In a possible implementation, the energy management server 501 is used to train the optimized model to obtain a trained power prediction model, including: the energy management server 501 is used to: obtain a training data set; the training data set includes a first historical meteorological data sequence of first meteorological parameters, a first historical photovoltaic power generation power data sequence, and a first timestamp data sequence; the first meteorological parameter is a meteorological parameter whose correlation with the photovoltaic power generation power is greater than a correlation threshold; perform clustering analysis on the training data set by using the Gaussian mixture model clustering method, and divide the training data set into a sunny day data set, a cloudy day data set, and a rainy day data set based on the clustering analysis result; generate sunny day samples, cloudy day samples, and rainy day samples respectively based on the sunny day data set, the cloudy day data set, and the rainy day data set; decompose each day sample into components with different frequencies by using the composite mode symmetric empirical mode decomposition adaptive noise algorithm; and train the optimized model by sequentially inputting each component of each day sample into the optimized model in an iterative manner to obtain a trained power prediction model; where, in the process of any iteration, after each component of the corresponding day sample is input into the optimized model, the predicted value of each component is obtained, and the power prediction value of the corresponding photovoltaic power generation is obtained by superimposing the predicted values of each component.

[0120] In a possible implementation, the coordination controller 502 can also be responsible for monitoring the real-time data and status information of local controllers such as the battery management system (BMS), programmable logic controller (PLC), and photovoltaic control unit (FCU), and performing data interaction and command execution with the energy management server 501.

[0121] In a possible implementation, the digital twin module has functions such as real-time display, fault warning, data statistical analysis, data reports, and permission management.

[0122] In a possible implementation, the substation data acquisition device and monitoring and control system are responsible for system data acquisition and monitoring, and engineers can perform system control and operation and maintenance management through the human-machine interface (HMI).

[0123] In a possible implementation, the energy management server 501 can receive the system data uplinked by the data acquisition gateway through the TCP / IP protocol and store it in the historical database. The system data can include meteorological data collected by a multi-factor meteorological instrument.

[0124] In a possible implementation, the energy management server 501 can run a prediction program developed based on the Python / TensorFlow architecture in real time and access the historical database to implement the power prediction function of photovoltaic power generation. The prediction results can call the Matlab script program through the matlab.engine module to generate the daily energy storage battery output plan. The output plan can be sent to the data acquisition gateway through the TCP / IP protocol and converted into the corresponding protocol and sent to the relevant device address, so as to achieve power request control. Engineers can access the historical database through the human-machine interface and modify the data.

[0125] See Figure 6 , Figure 6 This is a structural block diagram of an electronic device provided by an embodiment of the present application. As Figure 6 shown, the electronic device 600 may include a processor 601 and a memory 602; the memory 602 can be coupled to the processor 601. It should be noted that this Figure 6 is exemplary; other types of structures can also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0126] In a possible implementation, the functions of the energy management server 501 can be integrated into the processor 601.

[0127] In one possible implementation, the energy management server 501 can be configured separately from the processor 601. For example, the functions of the energy management server 501 can be configured as a chip connected to the processor 601, and energy management is achieved through the control of the processor 601.

[0128] In addition, in some alternative implementations, the electronic device 600 may further include: a communication module, an input unit, an audio processor, a display, a power supply, etc. It should be noted that the electronic device 600 does not necessarily have to include Figure 6 all the components shown in; in addition, the electronic device 600 may further include Figure 6 components not shown in, and reference can be made to the prior art.

[0129] In some alternative implementations, the processor 601 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor devices and / or logic devices. The processor 601 receives inputs and controls the operations of the various components of the electronic device 600.

[0130] Among them, the memory 602 can be, for example, one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices. It can store the information related to the energy management server 501 mentioned above. In addition, it can also store programs for executing relevant information. And the processor 601 can execute the programs stored in the memory 602 to implement information storage or processing, etc.

[0131] The input unit can provide inputs to the processor 601. The input unit is, for example, a key or a touch input device. The power supply can be used to supply power to the electronic device 600. The display can be used to display display objects such as images and texts. The display can be, for example, an LCD display, but is not limited thereto.

[0132] The memory 602 can be a solid-state memory. For example, a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that stores information even when powered off, can be selectively erased and has more data. Examples of such a memory are sometimes referred to as EPROMs, etc. The memory 602 can also be some other type of device. The memory 602 includes a buffer memory (sometimes referred to as a buffer). The memory 602 can include an application / function storage part, which is used to store application programs and function programs or the processes for operating the electronic device 1000 through the processor 601.

[0133] The memory 602 may also include a data storage unit for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit of the memory 602 may include various drivers of the electronic device for communication functions and / or for performing other functions of the electronic device (such as a messaging application, an address book application, etc.).

[0134] The communication module is a transmitter / receiver that transmits and receives signals via an antenna. The communication module (transmitter / receiver) is coupled to the processor 601 to provide input signals and receive output signals, which may be the same as in the case of a conventional mobile communication terminal.

[0135] Based on different communication technologies, multiple communication modules may be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module (transmitter / receiver) is also coupled to a speaker and a microphone via an audio processor to provide an audio output via the speaker and receive an audio input from the microphone, thereby implementing normal telecommunication functions. The audio processor may include any suitable buffers, decoders, amplifiers, etc. Additionally, the audio processor is also coupled to the processor 601, so that recording can be performed on the local device through the microphone, and the sound stored on the local device can be played through the speaker.

[0136] Embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps in the energy management method for an optical storage system in the above embodiments. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all the steps in the energy management method for the optical storage system in the above embodiments are implemented.

[0137] Although the present application provides method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on routine or non-creative labor. The order of steps listed in the embodiments is only one way among the execution orders of numerous steps and does not represent the only execution order. When the actual device or client product is executed, it may be executed in the order of the method shown in the embodiments or the drawings or in parallel (such as in an environment of parallel processors or multi-threaded processing).

[0138] Those skilled in the art should understand that the embodiments of this specification may be provided as a method, a device (system), or a computer program product. Therefore, the embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0140] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0141] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0142] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key point of each embodiment is to illustrate the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content. In this document, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances. It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. This application is not limited to any single aspect, nor to any single embodiment, nor to any arbitrary combination and / or permutation of these aspects and / or embodiments. Moreover, each aspect and / or embodiment of this application can be used alone or in combination with one or more other aspects and / or their embodiments.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, not to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of this application.

[0144] The above has described this application in combination with preferred embodiments, but these embodiments are only exemplary and only play an illustrative role. On this basis, various replacements and improvements can be made to this application, and these all fall within the protection scope of this application.

Claims

1. An energy management method for a photovoltaic storage system, characterized in that: The method comprises: Determine real number variables and binary variables for constructing an energy management optimization model for a photovoltaic storage system; the real number variables include the output of the energy storage battery, the state of charge of the energy storage battery, and the output of the power grid; the binary variables include the operating state of the energy storage battery, and the operating state of the energy storage battery is used to characterize whether the energy storage battery is in a charging state or a discharging state; Based on the real number variables and binary variables, an objective function is constructed with the goal of minimizing the total system cost and the energy storage battery life attenuation; the total system cost includes the grid electricity purchase and sales cost, photovoltaic power generation maintenance cost and energy storage battery attenuation loss cost; Determine optimization constraints based on the real number variables and binary variables; the optimization constraints include power balance constraints of the photovoltaic storage system, output constraints of the energy storage battery, and output constraints of the power grid; Based on the objective function and the optimization constraints, a non-convex mixed integer quadratic constrained nonlinear programming problem model is constructed as an energy management optimization model of the photovoltaic storage system; Obtaining a power prediction value of photovoltaic power generation in the photovoltaic storage system; Based on the power prediction value, optimizing and solving the energy management optimization model of the photovoltaic storage system to obtain the output plan of the energy storage battery in the photovoltaic storage system; The energy storage battery is operated and regulated based on the output plan.

2. The method according to claim 1, characterized in that Before constructing the objective function, the method further includes: According to the following formula, determine the service life of the energy storage battery after the i-th discharge; Among them, N life,i represents the service life of the energy storage battery after the i-th discharge; D od,i represents the discharge depth of the energy storage battery at the i-th discharge; α1, α2, α3, α4, α5 represent fitting parameters; i is a positive integer; Based on the service life of the energy storage battery after the i-th discharge, the loss cost of the energy storage battery for the i-th discharge is determined according to the following formula; Wherein, γ%(i) represents the loss cost of the energy storage battery during the i-th discharge.

3. The method according to claim 2, characterized in that The objective function includes the following relationship: C pv =ξ pv ×E pv ; Among them, C cost Represents the total cost of the system; C grid represents the power purchase and sale cost of the power grid; C pv represents the photovoltaic power generation maintenance cost; C bat represents the attenuation loss cost of the energy storage battery; t represents the tth sampling time; T represents the total number of samples; t and T are both positive integers; S buy (t) represents the electricity price that the photovoltaic storage system purchases from the power grid at the t-th sampling moment; S sell (t) represents the price of electricity sold by the photovoltaic storage system to the power grid at the t-th sampling moment; P grid represents the power grid output; b represents the unit cost of the energy storage battery; E bat represents the capacity of the energy storage battery; γ%(t) represents the loss cost of the energy storage battery discharge at the tth sampling moment; ξ pv Represents the photovoltaic power generation maintenance factor; E pv Represents photovoltaic installed capacity.

4. The method according to claim 3, characterized in that The power balance constraint condition of the photovoltaic storage system includes the following relationship: P pv +P bat +P grid =P L ; Among them, P pv Represents the power prediction value of photovoltaic power generation; P bat Indicates the output of the energy storage battery; P grid Indicates the power grid output; P L Indicates load power.

5. The method according to claim 3, characterized in that The energy storage battery output constraint condition includes the following relationship: Wherein, λ represents the operating state of the energy storage battery. When the energy storage battery is discharging, λ is equal to 1, and when the energy storage battery is charging, λ is equal to 0; P bat,min Indicates the lower limit of the energy storage battery output; P bat,max Indicates the upper limit of the energy storage battery output; P bat (:,k) represents the output of the energy storage battery at the kth sampling moment, where k is a positive integer.

6. The method according to claim 3, characterized in that The power grid output constraint condition includes the following relationship: P grid,min ≤P grid ≤P grid,max ; Among them, P grid,min Indicates the lower limit of the power grid output, P grid,max Indicates the upper limit of the power grid output.

7. The method according to any one of claims 1 to 6, characterized in that: The obtaining of the power prediction value of photovoltaic power generation in the photovoltaic storage system comprises: The second-generation non-dominated sorting genetic algorithm based on tent mapping was used to optimize the learning rate, number of neurons, and regularization coefficient of the pre-established convolutional neural network-bidirectional long short-term memory network-attention mechanism hybrid deep learning model to obtain the optimized model. Training the optimization model to obtain a trained power prediction model; The power prediction model is used to predict the power prediction value of photovoltaic power generation in the photovoltaic storage system.

8. The method according to claim 7, characterized in that Training the optimization model to obtain a trained power prediction model includes: Acquire a training data set; the training data set includes a first historical meteorological data sequence of a first meteorological parameter, a first historical photovoltaic power generation data sequence, and a first timestamp data sequence; the first meteorological parameter is a meteorological parameter whose correlation with the photovoltaic power generation power is greater than a correlation threshold; Performing cluster analysis on the training data set using a Gaussian mixture model clustering method, and dividing the training data set into a sunny day sub-data set, a cloudy day sub-data set, and a rainy day sub-data set based on the cluster analysis result; Generate a sunny day sample, a cloudy day sample, and a rainy day sample based on the sunny day sub-dataset, the cloudy day sub-dataset, and the rainy day sub-dataset, respectively; Each daily sample is decomposed into components of different frequencies using the composite mode symmetry empirical mode decomposition adaptive noise algorithm; In an iterative manner, each component of each daily sample is input into the optimization model in turn, and the optimization model is trained to obtain a trained power prediction model; Among them, in any iteration process, each component of the corresponding daily sample is input into the optimization model respectively to obtain the predicted value of each component, and the power prediction value of photovoltaic power generation corresponding to the corresponding daily sample is obtained based on the superposition of the predicted values ​​of each component.

9. An energy management system for a photovoltaic storage system, characterized in that: The system comprises: an energy management server; the energy management server is used for: Determine real number variables and binary variables for constructing an energy management optimization model for a photovoltaic storage system; the real number variables include the output of the energy storage battery, the state of charge of the energy storage battery, and the output of the power grid; the binary variables include the operating state of the energy storage battery, and the operating state of the energy storage battery is used to characterize whether the energy storage battery is in a charging state or a discharging state; Based on the real number variables and binary variables, an objective function is constructed with the goal of minimizing the total system cost and the energy storage battery life attenuation; the total system cost includes the grid electricity purchase and sales cost, photovoltaic power generation maintenance cost and energy storage battery attenuation loss cost; Determine optimization constraints based on the real number variables and binary variables; the optimization constraints include power balance constraints of the photovoltaic storage system, output constraints of the energy storage battery, and output constraints of the power grid; Based on the objective function and the optimization constraints, a non-convex mixed integer quadratic constrained nonlinear programming problem model is constructed as an energy management optimization model of the photovoltaic storage system; Obtaining a power prediction value of photovoltaic power generation in the photovoltaic storage system; Based on the power prediction value, optimizing and solving the energy management optimization model of the photovoltaic storage system to obtain the output plan of the energy storage battery in the photovoltaic storage system; The energy storage battery is operated and regulated based on the output plan.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 8 is implemented.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images