Solar power generation and energy storage system and method

Abstract

The present invention relates to the field of solar energy storage technology, and discloses a solar power generation and energy storage system and method, which comprises the following steps: collecting historical weather data, historical electricity consumption data, and user solar equipment data; analyzing the historical weather data to obtain an adjustment coefficient; analyzing the historical electricity consumption data to construct a user's electricity consumption behavior model, and predicting the user's electricity load data within the next 72 hours; obtaining a set of photovoltaic power generation within the next 72 hours based on weather forecast data within the next 72 hours provided by the meteorological department; constructing a charging and discharging decision optimization model, and executing the lowest-cost charging and discharging strategy among a group of charging and discharging strategies as the optimized charging and discharging strategy; by accurately predicting photovoltaic power generation and user electricity load, the present invention can optimize the charging and discharging strategy of the energy storage device, ensure that the self-generated electricity is used as much as possible, reduce dependence on the power grid, and reduce the user's electricity bill expenditure.

Description

Technical Field

[0001] The present invention relates to the field of solar energy storage, and more particularly, to a solar power generation and energy storage system and method. Background Art

[0002] Solar power storage involves storing the electricity generated by photovoltaic panels for use during periods of low sunlight, such as at night or on cloudy days. First, photovoltaic (PV) panels convert the sunlight they receive into direct current (DC) electricity. Since household and industrial electricity is typically alternating current (AC), an inverter is required to convert the DC power to AC. Excess energy is then transferred to storage batteries for storage via a charge controller. Modern energy storage systems may utilize various battery types, including lithium-ion, lead-acid, and sodium-sulfur. Lithium-ion batteries are preferred due to their high energy density, long lifespan, and low self-discharge rate. Advanced energy storage systems are typically equipped with an intelligent management system that monitors parameters such as charge level, health status, and temperature, optimizing charging and discharging strategies based on user consumption patterns and grid electricity prices. Energy storage systems can be designed to be connected to the grid (on-grid) or operate independently (off-grid). The grid-connected system can sell excess electricity back to the grid when the energy storage battery is fully charged, and purchase electricity from the grid to supplement the energy storage when the electricity price is low; the off-grid system relies entirely on its own power generation and energy storage; when the grid is out of power, the energy storage system can provide power support as an emergency backup power source.

[0003] However, the charging and discharging strategies in existing solar power generation and energy storage systems are usually executed according to fixed rules and cannot be optimized according to real-time conditions. As a result, the electricity costs of existing charging and discharging strategies are sometimes high and sometimes low, which is not conducive to reducing people's electricity costs. Summary of the Invention

[0004] The present invention provides a solar power generation and energy storage system and method to solve the technical problems raised in the above background technology.

[0005] The present invention provides the following steps:

[0006] Step 101, collecting historical weather data, historical electricity consumption data and user solar equipment data;

[0007] Step 102: Analyze historical weather data to obtain key factors affecting photovoltaic panel power generation efficiency. Combined with the weather characteristics of the region where the user's solar power generation equipment is located, the seasonal variation pattern of the region is obtained. Based on the seasonal variation pattern of the region, a corresponding adjustment coefficient is set for each season.

[0008] Step 103: Analyze historical electricity consumption data, build a user's electricity consumption behavior model, and predict the user's electricity load data within the next 72 hours;

[0009] Step 104: Based on the weather forecast data for the next 72 hours provided by the meteorological department, which includes hourly sunshine duration information, solar radiation intensity information, and ambient temperature information, the weather forecast data is input into the photovoltaic power generation model to obtain the predicted photovoltaic power generation for each hour in the future. The predicted photovoltaic power generation for each hour in the future is corrected according to the adjustment coefficient in step 102 to obtain the photovoltaic power generation set for the next 72 hours.

[0010] Step 105: Construct a charge-discharge decision optimization model. The predicted user's electric load data for the next 72 hours and the photovoltaic power generation value per hour for the next 72 hours are encoded and input into the charge-discharge decision optimization model. The charge-discharge decision optimization model outputs a set of charge-discharge strategies. Each charge-discharge strategy includes the total charge amount and total discharge amount of the solar energy storage device corresponding to each hour in the next 72 hours. The charge-discharge strategy with the lowest cost among the set of charge-discharge strategies is executed as the optimized charge-discharge strategy.

[0011] In a preferred embodiment, the historical weather data is the weather data of the area where the user's solar power generation equipment is located in the past three years, which includes hourly sunshine time information, solar radiation intensity information and ambient temperature information; the historical electricity consumption data is the user's electricity load data in the past three years, and the electricity load data includes peak period information, off-peak period information and total electricity consumption information; the user's solar equipment data includes a photovoltaic power generation model, which is used to calculate the hourly photovoltaic power generation based on the hourly sunshine time information, solar radiation intensity information and ambient temperature information.

[0012] In a preferred embodiment, the charge-discharge decision optimization model includes an encoding layer, a feature combination layer, a hidden layer, and an output layer;

[0013] The predicted user's electricity load data for the next 72 hours and the hourly photovoltaic power generation values ​​for the next 72 hours are encoded into photovoltaic power generation features and electricity consumption behavior features through a time embedding encoding method; the photovoltaic power generation features and electricity consumption behavior features are then input into the feature combination layer for feature fusion to generate a comprehensive feature; the comprehensive feature is then input into the hidden layer to output the hidden feature, and finally the hidden feature is input into the fully connected layer, which outputs a set of charging and discharging strategies.

[0014] In a preferred embodiment, the adjustment coefficient is set according to the seasonal variation law, and its calculation formula is:

[0015]

[0016] in, Indicates the adjustment coefficient set for different seasons, represents a seasonal variation function and is related to the season, average ambient temperature, and average solar radiation intensity. Indicates seasons, including spring, summer, autumn and winter, represents the average ambient temperature for a particular season, Represents the average solar radiation intensity in a specific season.

[0017] In a preferred embodiment, the calculation formula of the seasonal variation function is:

[0018]

[0019] in, represents the seasonal variation function, represents the average ambient temperature for a particular season, Indicates the temperature of solar power generation equipment under rated working conditions. represents the average solar radiation intensity in a particular season, Indicates the solar radiation intensity of solar power generation equipment under rated working conditions. 、 、 and represent the first, second, third and fourth empirical parameters respectively.

[0020] In a preferred embodiment, the calculation formula of the photovoltaic power generation model is:

[0021]

[0022] in, represents the photovoltaic power generation at time t, Indicates the effective area of ​​solar power generation equipment exposed to sunlight. represents the solar radiation intensity at time t, It represents the conversion efficiency function at the ambient temperature at time t and generally decreases with increasing temperature.

[0023] In a preferred embodiment, the calculation formula for the revised predicted future hourly photovoltaic power generation is:

[0024]

[0025] in, represents the revised predicted photovoltaic power generation per hour in the future, represents the photovoltaic power generation at time t, represents the adjustment coefficient at time t.

[0026] In a preferred embodiment, the method for constructing a charge-discharge decision optimization model includes the following steps:

[0027] Step 201, encoding input data;

[0028] First, the predicted user's electricity load data for the next 72 hours and the photovoltaic power generation value per hour in the next 72 hours are encoded;

[0029] The coding format of photovoltaic power generation is as follows: , t=1 to 72;

[0030] User's electricity load data: , t=1 to 72, represents the power demand of the user at time t;

[0031] Step 202, defining charge and discharge decision variables;

[0032] Two charging and discharging decision variables are defined to describe the behavior of the energy storage device:

[0033] represents the charge at time t;

[0034] Indicates the discharge amount at time t;

[0035] Step 203: Setting an objective function. The optimization goal is to find a charging and discharging strategy that minimizes the total cost, where the cost includes the cost of purchasing electricity from the grid and the operating cost of the energy storage device.

[0036] The objective function for minimizing the cost in calculating the charging and discharging strategy is:

[0037]

[0038] in, represents the total cost of the i-th charging and discharging strategy, represents the function that takes the minimum value, represents the electricity price at time t, represents the power demand of the user at time t, represents the photovoltaic power generation at time t, represents the discharge amount of the i-th charge and discharge strategy at time t, represents the charge capacity of the i-th charge-discharge strategy at time t, represents the operating cost of the energy storage equipment in the i-th charging and discharging strategy;

[0039] The calculation formula for the operating cost of energy storage equipment is:

[0040]

[0041] in, represents the operating cost of the energy storage equipment in the i-th charging and discharging strategy, represents the charging cost of the energy storage device at time t for the i-th charging and discharging strategy, represents the charging cost of the energy storage device at time t for the i-th charging and discharging strategy, represents the charging energy of the energy storage device at time t for the i-th charging and discharging strategy, represents the discharged energy of the energy storage device at time t for the i-th charge and discharge strategy;

[0042] Step 204: Set constraints, which include: one,

[0044]

[0045] in, At time t At time t ; two,

[0047]

[0048] in, Indicates the energy storage state, Indicates the maximum capacity; three,

[0050]

[0051]

[0052] in, 、 They represent the charge and discharge amounts at time t, 、 Represent the maximum power of charging and discharging respectively;

[0053] Step 205 : Obtain the lowest cost charging and discharging strategy according to a linear programming or mixed integer linear programming optimization algorithm.

[0054] A solar power generation and energy storage system includes the following modules:

[0055] Data collection module, used to collect historical weather data, historical electricity consumption data and user solar equipment data;

[0056] Weather analysis module, which is used to analyze historical weather data, determine the key factors affecting the efficiency of photovoltaic panels, and set adjustment coefficients based on seasonal changes;

[0057] The electricity consumption behavior modeling module is used to build a user electricity consumption behavior model based on historical electricity consumption data and predict electricity load information within the next 72 hours;

[0058] The photovoltaic power generation prediction module uses the weather forecast data for the next 72 hours provided by the meteorological department, combined with the photovoltaic power generation model to predict the future photovoltaic power generation and apply the adjustment coefficient for correction;

[0059] The charge and discharge decision optimization module is used to build and solve the charge and discharge decision optimization model and select the charge and discharge strategy with the lowest cost.

[0060] A storage medium stores non-transitory computer-readable instructions. When the non-transitory computer-readable instructions are executed by a computer, the steps in the above-mentioned solar power generation and energy storage method can be executed.

[0061] The beneficial effects of the present invention are:

[0062] 1. By accurately predicting photovoltaic power generation and user electricity load, the present invention can optimize the charging and discharging strategies of energy storage devices, ensuring that self-generated electricity is used as much as possible, reducing dependence on the power grid, charging when electricity prices are low, and discharging when electricity prices are high or photovoltaic output is insufficient, thereby reducing users' electricity bills;

[0063] 2. Based on historical data and real-time weather forecasts, the present invention enables the energy storage system to dynamically adjust its energy storage strategy and flexibly respond to changing environmental conditions and electricity demand. Even when photovoltaic output is low, such as on cloudy days or at night, the energy storage system can ensure a stable power supply and improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a flow chart of the solar power generation and energy storage method of the present invention.

[0065] Figure 2 It is a flow chart of the method for constructing a charge and discharge decision optimization model of the present invention. DETAILED DESCRIPTION

[0066] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. In addition, features described with respect to some examples may also be combined in other examples.

[0067] like Figure 1 As shown, a solar power generation and energy storage method includes the following steps:

[0068] Step 101: Collect historical weather data, historical electricity consumption data, and user solar power equipment data. The historical weather data refers to weather data for the area where the user's solar power generation equipment is located over the past three years, including hourly sunshine duration information, solar radiation intensity information, and ambient temperature information. The historical electricity consumption data refers to the user's electricity load data over the past three years, including peak period information, off-peak period information, and total electricity consumption information. The user's solar power equipment data includes a photovoltaic power generation model, which is used to calculate hourly photovoltaic power generation based on hourly sunshine duration information, solar radiation intensity information, and ambient temperature information.

[0069] It should be noted that user solar equipment data also includes various parameters of solar energy storage equipment and solar power generation equipment, such as rated power generation power, maximum storage capacity, and rated charge and discharge power.

[0070] In one embodiment of the present invention, the calculation formula of the photovoltaic power generation model is:

[0071]

[0072] in, represents the photovoltaic power generation at time t, Indicates the effective area of ​​solar power generation equipment exposed to sunlight. represents the solar radiation intensity at time t, It represents the conversion efficiency function at the ambient temperature at time t and generally decreases with increasing temperature.

[0073] Step 102: Analyze historical weather data to obtain key factors affecting photovoltaic panel power generation efficiency. Combined with the weather characteristics of the region where the user's solar power generation equipment is located, the seasonal variation pattern of the region is obtained. Based on the seasonal variation pattern of the region, a corresponding adjustment coefficient is set for each season.

[0074] It should be noted that the key factors affecting the power generation efficiency of photovoltaic panels include temperature and radiation intensity;

[0075] In one embodiment of the present invention, the adjustment coefficient is set according to the seasonal variation law, as shown below:

[0076]

[0077] in, Indicates the adjustment coefficient set for different seasons, represents a seasonal variation function and is related to the season, average ambient temperature, and average solar radiation intensity. Indicates seasons, including spring, summer, autumn and winter, represents the average ambient temperature for a particular season, Indicates the average solar radiation intensity in a particular season;

[0078] The calculation formula of the seasonal variation function is:

[0079]

[0080] in, represents the seasonal variation function, represents the average ambient temperature for a particular season, Indicates the temperature of solar power generation equipment under rated working conditions. represents the average solar radiation intensity in a particular season, Indicates the solar radiation intensity of solar power generation equipment under rated working conditions. 、 、 and denote the first, second, third and fourth empirical parameters respectively;

[0081] It should be noted that and All of these can be found in the operating manual of the solar power generation equipment or on the product nameplate. 、 、 and These parameters can also be determined based on experimental data or information provided by the manufacturer. These parameters will determine the degree to which temperature and radiation intensity affect power generation efficiency.

[0082] Step 103: Analyze historical electricity consumption data, build a user's electricity consumption behavior model, and predict the user's electricity load data within the next 72 hours;

[0083] In one embodiment of the present invention, a method for constructing a user's electricity usage behavior model includes the following steps:

[0084] Step 1: Perform feature engineering on the collected historical electricity consumption data;

[0085] Time features: extracting daily, weekly, and monthly periodic patterns, as well as information such as weekdays / weekends and holidays;

[0086] External factors: Consider the impact of temperature and humidity meteorological conditions and electricity price changes on electricity consumption behavior;

[0087] Aggregate statistics: Calculate the total daily and weekly electricity consumption, and the average electricity consumption during peak and off-peak hours;

[0088] Step 2: Select the LSTM prediction model as the user's electricity consumption behavior model, divide the historical data into a training set and a validation set, and adjust the hyperparameters of the LSTM prediction model using the Bayesian optimization method to find the optimal configuration.

[0089] Step 3: Use the mean square error (MSE) indicator to evaluate the accuracy of the model, and use k-fold cross validation to evaluate the stability and generalization ability of the model;

[0090] Step 4: Input the weather forecast data for the next 72 hours provided by the meteorological department and the electricity price change information for the next 72 hours into the trained LSTM prediction model, predict the hourly electricity load for the next 72 hours, and obtain the user's electricity load data for the next 72 hours.

[0091] Step 104: Based on the weather forecast data for the next 72 hours provided by the meteorological department, which includes hourly sunshine duration information, solar radiation intensity information, and ambient temperature information, the weather forecast data is input into the photovoltaic power generation model to obtain the predicted photovoltaic power generation for each hour in the future. The predicted photovoltaic power generation for each hour in the future is corrected according to the adjustment coefficient in step 102 to obtain the photovoltaic power generation set for the next 72 hours.

[0092] It should be noted that the photovoltaic power generation set for the next 72 hours is a set of photovoltaic power generation values ​​for each hour in the next 72 hours arranged in chronological order, and includes 72 sets of photovoltaic power generation values. In other words, each photovoltaic power generation value corresponds to the photovoltaic power generation for a certain hour in the next 72 hours.

[0093] In one embodiment of the present invention, the calculation formula for the revised predicted future hourly photovoltaic power generation is:

[0094]

[0095] in, represents the revised predicted photovoltaic power generation per hour in the future, represents the photovoltaic power generation at time t, represents the adjustment coefficient at time t;

[0096] It should be noted that the time t in the present invention is the time period between the t-1th time point and the tth time point, and the length of the time period is usually expressed as 1 hour;

[0097] Step 105: Construct a charge-discharge decision optimization model. The predicted user's electric load data for the next 72 hours and the hourly photovoltaic power generation value for the next 72 hours are encoded and input into the charge-discharge decision optimization model. The charge-discharge decision optimization model outputs a set of charge-discharge strategies. Each charge-discharge strategy includes the total charge and discharge amount of the solar energy storage device corresponding to each hour in the next 72 hours. The charge-discharge strategy with the lowest cost among the set of charge-discharge strategies is executed as the optimized charge-discharge strategy.

[0098] The charge-discharge decision optimization model includes an encoding layer, a feature combination layer, a hidden layer, and an output layer;

[0099] The predicted user's electricity load data for the next 72 hours and the hourly photovoltaic power generation values ​​for the next 72 hours are encoded into photovoltaic power generation features and electricity consumption behavior features through a time embedding encoding method; the photovoltaic power generation features and electricity consumption behavior features are then input into the feature combination layer for feature fusion to generate a comprehensive feature; the comprehensive feature is then input into the hidden layer to output the hidden feature, and finally the hidden feature is input into the fully connected layer, which outputs a set of charging and discharging strategies.

[0100] It should be noted that the mth element in the comprehensive feature includes both the power generation of the photovoltaic power generation equipment at the mth hour and the power consumption of the user at the mth hour. In other words, the feature combination layer combines the photovoltaic power generation and the user's power consumption at each time in chronological order.

[0101] In one embodiment of the present invention, Figure 2 As shown, the method for constructing a charge and discharge decision optimization model includes the following steps:

[0102] Step 201, encoding input data;

[0103] First, the predicted user's electricity load data for the next 72 hours and the photovoltaic power generation value per hour in the next 72 hours are encoded;

[0104] The coding format of photovoltaic power generation is as follows: , t=1 to 72;

[0105] User's electricity load data: , t=1 to 72, represents the power demand of the user at time t;

[0106] Step 202, defining charge and discharge decision variables;

[0107] Two charging and discharging decision variables are defined to describe the behavior of the energy storage device:

[0108] represents the charge amount at time t (when the energy storage device is charged from the grid or photovoltaic system);

[0109] Represents the discharge amount at time t (when the energy storage device supplies power to the load);

[0110] Step 203: setting an objective function, the goal is to find a charging and discharging strategy that minimizes the total cost, where the cost includes the cost of purchasing electricity from the grid and the operating cost of the energy storage device;

[0111] The objective function for minimizing the cost in calculating the charging and discharging strategy is:

[0112]

[0113] in, represents the total cost of the i-th charging and discharging strategy, represents the function that takes the minimum value, represents the electricity price at time t, represents the power demand of the user at time t, represents the photovoltaic power generation at time t, represents the discharge amount of the i-th charge and discharge strategy at time t, represents the charge capacity of the i-th charge-discharge strategy at time t, represents the operating cost of the energy storage equipment in the i-th charging and discharging strategy;

[0114] The calculation formula for the operating cost of energy storage equipment is:

[0115]

[0116] in, represents the operating cost of the energy storage equipment in the i-th charging and discharging strategy, represents the charging cost of the energy storage device at time t for the i-th charging and discharging strategy, represents the charging cost of the energy storage device at time t for the i-th charging and discharging strategy, represents the charging energy of the energy storage device at time t for the i-th charging and discharging strategy, represents the discharged energy of the energy storage device at time t for the i-th charge and discharge strategy;

[0117] Step 204: Set constraints, which include: one,

[0119]

[0120] in, represents the user's electricity load data at time t, represents the photovoltaic power generation at time t, represents the discharge amount at time t, represents the charge at time t; two,

[0122]

[0123] in, Indicates the energy storage state, Indicates the maximum capacity; three,

[0125]

[0126]

[0127] in, 、 They represent the charge and discharge amounts at time t, 、 Represent the maximum power of charging and discharging respectively;

[0128] Step 205 : Obtain the lowest cost charging and discharging strategy according to a linear programming or mixed integer linear programming optimization algorithm.

[0129] A solar power generation and energy storage system includes the following modules:

[0130] Data collection module, used to collect historical weather data, historical electricity consumption data and user solar equipment data;

[0131] Weather analysis module, which is used to analyze historical weather data, determine the key factors affecting the efficiency of photovoltaic panels, and set adjustment coefficients based on seasonal changes;

[0132] The electricity consumption behavior modeling module is used to build a user electricity consumption behavior model based on historical electricity consumption data and predict electricity load information within the next 72 hours;

[0133] The photovoltaic power generation prediction module uses the weather forecast data for the next 72 hours provided by the meteorological department, combined with the photovoltaic power generation model to predict the future photovoltaic power generation and apply the adjustment coefficient for correction;

[0134] The charge and discharge decision optimization module is used to build and solve the charge and discharge decision optimization model and select the charge and discharge strategy with the lowest cost.

[0135] A storage medium stores non-transitory computer-readable instructions. When the non-transitory computer-readable instructions are executed by a computer, the steps in the above-mentioned solar power generation and energy storage method can be executed.

[0136] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation methods. The above specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A solar power generation and energy storage method, characterized in that: The following steps are involved: Step 101, collecting historical weather data, historical electricity consumption data and user solar equipment data; Step 102: Analyze historical weather data to obtain key factors affecting photovoltaic panel power generation efficiency. Combined with the weather characteristics of the region where the user's solar power generation equipment is located, the seasonal variation pattern of the region is obtained. Based on the seasonal variation pattern of the region, a corresponding adjustment coefficient is set for each season. Step 103: Analyze historical electricity consumption data, build a user's electricity consumption behavior model, and predict the user's electricity load data within the next 72 hours; Step 104: Based on the weather forecast data for the next 72 hours provided by the meteorological department, which includes hourly sunshine duration information, solar radiation intensity information, and ambient temperature information, the weather forecast data is input into the photovoltaic power generation model to obtain the predicted photovoltaic power generation for each hour in the future. The predicted photovoltaic power generation for each hour in the future is corrected according to the adjustment coefficient in step 102 to obtain the photovoltaic power generation set for the next 72 hours. Step 105: Construct a charge-discharge decision optimization model. The predicted user's electric load data for the next 72 hours and the hourly photovoltaic power generation value for the next 72 hours are encoded and input into the charge-discharge decision optimization model. The charge-discharge decision optimization model outputs a set of charge-discharge strategies. Each charge-discharge strategy includes the total charge and discharge amount of the solar energy storage device corresponding to each hour in the next 72 hours. The charge-discharge strategy with the lowest cost among the set of charge-discharge strategies is executed as the optimized charge-discharge strategy. The method for constructing a charge-discharge decision optimization model includes the following steps: Step 201, encoding input data; First, the predicted user's electricity load data for the next 72 hours and the photovoltaic power generation value per hour in the next 72 hours are encoded; The coding format of photovoltaic power generation is as follows: P ’ (t), t = 1 to 72; User's electricity load data: D(t), t = 1 to 72, represents the user's demand for electricity at time t; Step 202, defining charge and discharge decision variables; Two charging and discharging decision variables are defined to describe the behavior of the energy storage device: C(t) represents the charge at time t; D s (t) represents the discharge amount at time t; Step 203: Setting an objective function. The optimization goal is to find a charging and discharging strategy that minimizes the total cost, where the cost includes the cost of purchasing electricity from the grid and the operating cost of the energy storage device. The objective function for minimizing the cost in calculating the charging and discharging strategy is: Among them, y i represents the total cost of the i-th charging and discharging strategy, min represents the function of taking the minimum value, C grid (t) represents the electricity price at time t, D(t) represents the electricity demand of the user at time t, P ’ (t) represents the photovoltaic power generation at time t, represents the discharge capacity of the i-th charge and discharge strategy at time t, C i (t) represents the charge capacity of the i-th charge-discharge strategy at time t, represents the operating cost of the energy storage equipment in the i-th charging and discharging strategy; The calculation formula for the operating cost of energy storage equipment is: in, represents the operating cost of the energy storage equipment in the i-th charging and discharging strategy, represents the charging cost of the energy storage device at time t for the i-th charging and discharging strategy, represents the charging cost of the energy storage device at time t for the i-th charging and discharging strategy, C i (t) represents the charging energy of the energy storage device at time t for the i-th charging and discharging strategy, (t) represents the discharged energy of the energy storage device at time t for the i-th charge-discharge strategy; Step 204: Set constraints, which include: one, P’(t)+C(t)=D(t)+D s (t) Where D(t) represents the user's electricity load data at time t, P'(t) represents the photovoltaic power generation at time t, and D s (t) represents the discharge capacity at time t, and C(t) represents the charge capacity at time t; two, 0≤SOC(t)≤SOC max Among them, SOC represents the state of energy storage, SOC max Indicates the maximum capacity; three, 0≤C(t)≤C max 0≤D s (t)≤D s，max Among them, C(t), D s (t) represents the charge and discharge capacity at time t, C max 、D s，max Represent the maximum power of charging and discharging respectively; Step 205 : Obtain the lowest cost charging and discharging strategy according to a linear programming or mixed integer linear programming optimization algorithm.

2. A solar power generation and energy storage method according to claim 1, characterized in that: The historical weather data refers to the weather data of the area where the user's solar power generation equipment is located in the past three years, which includes hourly sunshine time information, solar radiation intensity information and ambient temperature information. The historical electricity consumption data refers to the user's electricity load data in the past three years, which includes peak period information, off-peak period information and total electricity consumption information; the user's solar equipment data includes a photovoltaic power generation model, which is used to calculate the hourly photovoltaic power generation based on the hourly sunshine time information, solar radiation intensity information and ambient temperature information.

3. A solar power generation and energy storage method according to claim 2, characterized in that: The charge-discharge decision optimization model includes an encoding layer, a feature combination layer, a hidden layer, and an output layer; The predicted user's electricity load data for the next 72 hours and the hourly photovoltaic power generation values ​​for the next 72 hours are encoded into photovoltaic power generation features and electricity consumption behavior features through a time embedding encoding method; the photovoltaic power generation features and electricity consumption behavior features are then input into the feature combination layer for feature fusion to generate a comprehensive feature; the comprehensive feature is then input into the hidden layer to output the hidden feature, and finally the hidden feature is input into the fully connected layer, which outputs a set of charging and discharging strategies.

4. A solar power generation and energy storage method according to claim 3, characterized in that: The adjustment coefficient is set according to the seasonal variation law, and its calculation formula is: K season =f(Season,T avg ,I avg ) Among them, K season represents the adjustment coefficient set for different seasons, f represents the seasonal variation function and is related to the season, average ambient temperature and average solar radiation intensity, Season represents the season, including spring, summer, autumn and winter, T avg represents the average ambient temperature in a particular season, I avg Represents the average solar radiation intensity in a specific season.

5. A solar power generation and energy storage method according to claim 4, characterized in that: The calculation formula of the seasonal variation function is: Where f represents the seasonal variation function, T avg represents the average ambient temperature in a particular season, T ref Indicates the temperature of solar power generation equipment under rated working conditions, I avg Indicates the average solar radiation intensity in a particular season, I ref represents the solar radiation intensity of the solar power generation equipment under rated working conditions, and a, b, c and d represent the first, second, third and fourth empirical parameters respectively.

6. A solar power generation and energy storage method according to claim 5, characterized in that: The calculation formula of the photovoltaic power generation model is: P(t)=A×I(t)×η(T(t)) Where P(t) represents the photovoltaic power generation at time t, A represents the effective area of ​​the solar power generation equipment exposed to sunlight, I(t) represents the solar radiation intensity at time t, and η(T(t)) represents the conversion efficiency function at the ambient temperature at time t, which generally decreases with increasing temperature.

7. A solar power generation and energy storage method according to claim 6, characterized in that: The calculation formula for the revised predicted future hourly photovoltaic power generation is: P ’ (t)=P(t)×K season (t) Among them, P ’ (t) represents the revised predicted photovoltaic power generation per hour in the future, P(t) represents the photovoltaic power generation at time t, K season (t) represents the adjustment coefficient at time t.

8. A solar power generation and energy storage system, used to implement a solar power generation and energy storage method according to any one of claims 1 to 7, characterized in that: Includes the following modules: Data collection module, used to collect historical weather data, historical electricity consumption data and user solar equipment data; Weather analysis module, which is used to analyze historical weather data, determine the key factors affecting the efficiency of photovoltaic panels, and set adjustment coefficients based on seasonal changes; The electricity consumption behavior modeling module is used to build a user electricity consumption behavior model based on historical electricity consumption data and predict electricity load information within the next 72 hours; The photovoltaic power generation prediction module uses the weather forecast data for the next 72 hours provided by the meteorological department, combined with the photovoltaic power generation model to predict the future photovoltaic power generation and apply the adjustment coefficient for correction; The charge and discharge decision optimization module is used to build and solve the charge and discharge decision optimization model and select the charge and discharge strategy with the lowest cost.

9. A storage medium, characterized in that: It stores non-transitory computer-readable instructions. When the non-transitory computer-readable instructions are executed by a computer, the steps in the solar power generation and energy storage method as claimed in claim 7 can be executed.

Images