Method and device for generating energy scheduling strategy based on photovoltaic energy storage equipment

Abstract

The invention provides a method and device for generating an energy scheduling strategy based on photovoltaic energy storage equipment, and the method comprises the steps: collecting peak and valley electricity price data and power utilization curve data of a to-be-tested factory, and carrying out the preprocessing, and obtaining target power utilization curve data and target peak and valley electricity price data; inputting the target electricity utilization curve data and the target peak-valley electricity price data into the large model for prediction to obtain a prediction result; and if the prediction result indicates that the to-be-tested factory is configured with the photovoltaic energy storage equipment, generating an energy scheduling strategy based on the prediction result, the target power utilization curve data and the target peak-valley electricity price data and outputting the energy scheduling strategy. By accurately predicting the condition of the photovoltaic energy storage equipment, an energy scheduling strategy is generated, energy management is optimized, and the power utilization cost is reduced. And the energy utilization efficiency of factories is improved. The system has strong universality and adaptability, can be flexibly adjusted according to the requirements of different factories, provides accurate and intelligent energy management and optimization for factories, assists energy conservation and emission reduction, and improves the overall management level.

Description

Technical Field

[0001] The present invention relates to the field of power systems and industrial automation technology, and in particular to a method and device for generating an energy scheduling strategy based on photovoltaic energy storage equipment. Background Art

[0002] As industrialization accelerates, factory electricity consumption increases, making energy costs a significant expense. The application of photovoltaic and energy storage technologies has garnered significant attention. PV systems convert solar energy into electricity, reducing reliance on the grid. Energy storage systems store energy during low electricity prices and release it during peak periods. This helps factories optimize their electricity usage, reduce electricity costs, and minimize carbon emissions, ultimately promoting green and sustainable development.

[0003] However, existing power management technologies primarily rely on traditional data analysis methods, which have limitations in assessing the effectiveness of photovoltaic and energy storage equipment and optimizing power usage. Traditional methods struggle to accurately assess whether a factory has installed photovoltaic and energy storage equipment, hindering more refined power scheduling and management. Furthermore, existing technologies lack adaptability and versatility, and lack the flexibility to adapt to the needs of individual factories, limiting their effectiveness in different scenarios.

[0004] Therefore, improving the intelligence level of energy management systems and developing technologies that can accurately predict equipment operation conditions and generate customized energy scheduling strategies have become technical challenges that need to be solved urgently. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a method and apparatus for generating an energy scheduling strategy based on photovoltaic energy storage equipment to solve the problem of difficult and low-precision energy scheduling.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] A first aspect of the present invention discloses a method for generating an energy scheduling strategy based on a photovoltaic energy storage device, the method comprising:

[0008] Collect peak and valley electricity price data and the power consumption curve data of the factory to be tested and pre-process them to obtain target power consumption curve data and target peak and valley electricity price data;

[0009] Inputting the target electricity consumption curve data and the target peak-valley electricity price data into a large model for prediction to obtain a prediction result; the large model is pre-trained based on historical electricity consumption curve data;

[0010] If the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment, an energy scheduling strategy is generated and output based on the prediction result, the target electricity consumption curve data and the target peak-valley electricity price data.

[0011] Preferably, the process of pre-training a large model based on historical electricity consumption curve data includes:

[0012] Collecting historical peak-valley electricity price data and historical electricity consumption curve data, and preprocessing the historical peak-valley electricity price data and the historical electricity consumption curve data to obtain a sample data set;

[0013] Dividing the sample data set into a training set and a test set;

[0014] Constructing an initial large model, and inputting the training set into the initial large model to obtain a prediction result corresponding to the training set;

[0015] Calculating the loss function of the initial large model according to the prediction result;

[0016] Determine whether the loss function meets a preset value;

[0017] If the loss function satisfies a preset value, the initial large model is optimized using the test set to obtain a final large model;

[0018] If the loss function does not meet the preset value, the parameters of the initial large model are updated according to the adaptive moment estimation optimization algorithm, and the step of calculating the loss function of the initial large model according to the prediction result is returned to be executed.

[0019] Preferably, the collecting of peak and valley electricity price data and electricity consumption curve data of the tested factory and preprocessing thereof to obtain target electricity consumption curve data and target peak and valley electricity price data includes:

[0020] Collect peak and valley electricity price data within a preset time period;

[0021] Collect power and electricity consumption information within a preset time period from the power monitoring system and electricity meter equipment of the factory to be tested, and obtain electricity consumption curve data;

[0022] The peak-valley electricity price data and the electricity consumption curve data are preprocessed to obtain target electricity consumption curve data and target peak-valley electricity price data.

[0023] Preferably, generating and outputting an energy scheduling strategy based on the prediction result, the target electricity consumption curve data and the peak-valley electricity price includes:

[0024] Extracting the photovoltaic installed capacity value and the energy storage battery capacity value from the prediction results;

[0025] Obtain grid power supply information and preset time periods;

[0026] Generate an energy scheduling strategy based on the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data and the peak and valley electricity price;

[0027] The energy scheduling strategy is output.

[0028] Preferably, the generating of the energy scheduling strategy based on the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data and the peak-valley electricity price includes:

[0029] Constructing an objective function according to the preset time period, the peak-valley electricity price and the target electricity consumption curve data;

[0030] Determining constraint conditions based on the target power consumption curve data, the photovoltaic installed capacity value, the energy storage battery capacity value, and the grid power supply information;

[0031] An energy scheduling strategy is generated according to a preset optimization algorithm, the objective function and the constraint conditions.

[0032] Preferably, after obtaining the power grid power supply information and the preset time period, the method further includes:

[0033] Obtain weather forecast data and production plans;

[0034] Generate an energy scheduling strategy based on the weather forecast data, the production plan, the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data and the peak and valley electricity price;

[0035] The energy scheduling strategy is output.

[0036] A second aspect of the present invention discloses a device for generating an energy scheduling strategy based on a photovoltaic energy storage device, the device comprising:

[0037] A target data collection unit is used to collect peak and valley electricity price data and electricity consumption curve data of the factory to be tested and pre-process them to obtain target electricity consumption curve data and target peak and valley electricity price data;

[0038] A large model prediction unit is used to input the target electricity consumption curve data and the target peak and valley electricity price data into the large model for prediction to obtain a prediction result; the large model is pre-trained based on historical electricity consumption curve data;

[0039] A generating unit is configured to generate and output an energy scheduling strategy based on the prediction result, the target electricity consumption curve data, and the target peak-valley electricity price data if the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment.

[0040] Preferably, the device further comprises:

[0041] a sample data collection unit, configured to collect historical peak-valley electricity price data and historical electricity consumption curve data, and pre-process the historical peak-valley electricity price data and the historical electricity consumption curve data to obtain a sample data set;

[0042] A division unit, configured to divide the sample data set into a training set and a test set;

[0043] A construction unit, configured to construct an initial large model, and input the training set into the initial large model to obtain a prediction result corresponding to the training set;

[0044] A calculation unit, configured to calculate a loss function of the initial large model according to the prediction result;

[0045] A judgment unit, configured to judge whether the loss function satisfies a preset value;

[0046] an optimization unit, configured to optimize the initial large model using the test set to obtain a final large model if the loss function satisfies a preset value;

[0047] An updating unit is used to update the parameters of the initial large model according to an adaptive moment estimation optimization algorithm if the loss function does not meet a preset value, and return to execute the calculation unit.

[0048] Preferably, the target data collection unit includes:

[0049] The first collection module is used to collect peak and valley electricity price data within a preset time period;

[0050] The second collection module is used to collect power and electricity consumption information within a preset time period from the power monitoring system and electricity meter equipment of the factory to be tested, and obtain electricity consumption curve data;

[0051] The preprocessing module is used to preprocess the peak-valley electricity price data and the electricity consumption curve data to obtain target electricity consumption curve data and target peak-valley electricity price data.

[0052] Preferably, the generating unit includes:

[0053] An extraction module, configured to extract the photovoltaic installed capacity value and the energy storage battery capacity value from the prediction result;

[0054] An acquisition module is used to obtain power grid power supply information and preset time periods;

[0055] A generation module, configured to generate an energy scheduling strategy based on the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data, and the peak and valley electricity price;

[0056] The output module is used to output the energy scheduling strategy.

[0057] Based on the above-mentioned embodiment of the present invention, a method and device for generating an energy scheduling strategy based on photovoltaic energy storage equipment is provided. Peak and valley electricity price data and the electricity consumption curve data of the factory to be tested are collected and preprocessed to obtain target electricity consumption curve data and target peak and valley electricity price data; the target electricity consumption curve data and target peak and valley electricity price data are input into a large model for prediction to obtain a prediction result; if the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment, an energy scheduling strategy is generated based on the prediction result, the target electricity consumption curve data and the target peak and valley electricity price data and output. By accurately predicting the status of photovoltaic energy storage equipment, an energy scheduling strategy is generated, energy management is optimized, and electricity costs are reduced. The energy utilization efficiency of the factory is improved. The present invention has strong versatility and adaptability, and can be flexibly adjusted according to the needs of different factories, providing factories with accurate and intelligent energy management and optimization, helping to save energy and reduce emissions, and improve the overall management level. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0059] Figure 1 A flowchart of a method for generating an energy scheduling strategy based on photovoltaic energy storage equipment provided by an embodiment of the present invention;

[0060] Figure 2 An example diagram comparing electricity costs of a factory under test provided in an embodiment of the present invention;

[0061] Figure 3 A structural block diagram of an apparatus for generating an energy scheduling strategy based on photovoltaic energy storage equipment provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0063] In this application, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0064] As can be seen from the background technology, existing power management technology relies on traditional data analysis methods, which makes it difficult to accurately evaluate the effects of photovoltaic and energy storage equipment, and lacks flexibility, which limits fine scheduling.

[0065] Therefore, an embodiment of the present invention provides a method and apparatus for generating an energy scheduling strategy based on photovoltaic energy storage equipment, collecting peak and valley electricity price data and the electricity consumption curve data of the factory to be tested and performing preprocessing to obtain target electricity consumption curve data and target peak and valley electricity price data; inputting the target electricity consumption curve data and target peak and valley electricity price data into a large model for prediction to obtain a prediction result; if the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment, then generating an energy scheduling strategy based on the prediction result, target electricity consumption curve data and target peak and valley electricity price data and outputting it. By accurately predicting the status of photovoltaic energy storage equipment, an energy scheduling strategy is generated, energy management is optimized, and electricity costs are reduced. The energy utilization efficiency of the factory is improved. The present invention has strong versatility and adaptability, and can be flexibly adjusted according to the needs of different factories, providing factories with accurate and intelligent energy management and optimization, helping to save energy and reduce emissions, and improve the overall management level.

[0066] See also Figure 1 , shows a flow chart of a method for generating an energy scheduling strategy based on photovoltaic energy storage equipment provided by an embodiment of the present invention. The method includes:

[0067] Step S101: collecting peak and valley electricity price data and electricity consumption curve data of the factory to be tested and preprocessing them to obtain target electricity consumption curve data and target peak and valley electricity price data.

[0068] In the specific implementation of step S101, peak-valley electricity price data is obtained from the power grid and electricity consumption curve data is collected from the factory to be tested. The peak-valley electricity price data and electricity consumption curve data are preprocessed to obtain target electricity consumption curve data and target peak-valley electricity price data.

[0069] Specifically, it includes the following processes (process A1 to process A3):

[0070] Process A1: Collect peak and valley electricity price data within a preset time period.

[0071] In the specific implementation process A1, peak and valley electricity price data within a preset time period are collected from the power grid, including peak period electricity price data, valley period electricity price data and flat period electricity price data.

[0072] For example, obtain peak and valley electricity price data for each period of the past month from the power grid, including peak hours (10:00-15:00 and 18:00-23:00 every day), valley hours (0:00-7:00 every day), and flat hours (7:00-10:00, 15:00-18:00, and 23:00-24:00 every day).

[0073] Process A2: Collect power and electricity consumption information within a preset time period from the power monitoring system and electricity meter equipment of the factory to be tested, and obtain electricity consumption curve data.

[0074] When specifically implementing process A2, the power and electricity consumption information within a preset time period is collected from the power monitoring system and electricity meter equipment of the factory to be tested, and electricity consumption curve data is constructed.

[0075] For example, collect hourly power and electricity consumption information for the past month from the power monitoring system and meter equipment of the factory to be tested, and record the corresponding date and time.

[0076] Process A3: Preprocess the peak-valley electricity price data and the electricity consumption curve data to obtain target electricity consumption curve data and target peak-valley electricity price data.

[0077] When specifically implementing process A3, preprocessing operations such as data cleaning, missing value filling, and data standardization are performed on the peak-valley electricity price data and the electricity consumption curve data to obtain the target electricity consumption curve data and the target peak-valley electricity price data.

[0078] As you can understand, preprocessing specifically involves cleaning the peak and valley electricity price data and electricity consumption curve data to remove outliers and erroneous data. Missing values ​​are filled using linear interpolation. Next, the peak and valley electricity price data and electricity consumption curve data are normalized and converted to the [0, 1] interval to improve data quality and usability, facilitating subsequent processing of large models.

[0079] Step S102: Input the target electricity consumption curve data and the target peak-valley electricity price data into the large model for prediction to obtain a prediction result.

[0080] In the specific implementation of step S102 , the target electricity consumption curve data and the target peak-valley electricity price data obtained after preprocessing are input into the large model, and the large model is used to perform prediction to obtain a prediction result.

[0081] It should be noted that the prediction results include the probability of the tested factory installing photovoltaic equipment and the probability of installing energy storage equipment.

[0082] For plants equipped with photovoltaic and energy storage systems, the forecast results can also include predicted values ​​for photovoltaic installed capacity and energy storage battery capacity, providing a basis for subsequent energy scheduling strategies. By optimizing the parameters of the large model, it can accurately predict the status of the plant's photovoltaic and energy storage equipment and related parameters.

[0083] It is understandable that the large model is pre-trained based on historical electricity consumption curve data. The specific training process is as follows (process B1 to process B7):

[0084] Process B1: Collect historical peak-valley electricity price data and historical electricity consumption curve data, pre-process the historical peak-valley electricity price data and historical electricity consumption curve data, and obtain a sample data set.

[0085] In the specific implementation process B1, historical peak and valley electricity price data are collected from the power grid and historical electricity consumption curve data (for example, historical peak and valley electricity price data and historical electricity consumption curve data for the past six months) are collected from the power monitoring system and electricity meter equipment of the factory to be tested. The historical peak and valley electricity price data and historical electricity consumption curve data are preprocessed to obtain a sample data set.

[0086] It should be noted that the specific implementation principle of collecting historical peak and valley electricity price data and historical electricity consumption curve data and preprocessing is consistent with the specific implementation principle of collecting peak and valley electricity price data and electricity consumption curve data and preprocessing in the above step S101, and will not be repeated here.

[0087] Step B2: Divide the sample dataset into training set and test set.

[0088] When specifically implementing process B2, the sample data set obtained by preprocessing is divided into two parts: a training set and a test set.

[0089] For example, the sample data set is divided into a training set and a test set in a ratio of 8:2.

[0090] Process B3: Build an initial large model and input the training set into the initial large model to obtain the prediction results corresponding to the training set.

[0091] When implementing process B3, select a large model architecture (such as a recurrent neural network (RNN), a long short-term memory network (LSTM), or a variational autoencoder (VAE)) to build an initial large model. Then, input the training set into the initial large model and use it to perform predictions, obtaining prediction results corresponding to the training set.

[0092] The following uses the Long Short-Term Memory (LSTM) network as an example to illustrate how to build the initial large model.

[0093] First, a neural network was constructed consisting of two layers of LSTM units, each with 128 neurons. The network was terminated with a fully connected layer with a dimensionality of 3, representing the presence of photovoltaic (PV) and energy storage equipment installed in the factory under test, as well as relevant parameters of the PV and energy storage systems (e.g., predicted PV installed capacity and battery capacity).

[0094] For example, based on a set threshold (such as 0.5), if the first output dimension value is greater than the threshold, it indicates that the factory under test has installed photovoltaic equipment; if the second output dimension value is greater than the threshold, it indicates that the factory under test has installed energy storage equipment.

[0095] Secondly, the learning rate was set to 0.001. During the training process, the historical electricity consumption curve data and historical peak and valley electricity price data in the training set were used as input features, and the probability of the tested factory installing photovoltaic equipment and the probability of installing energy storage equipment were used as output labels.

[0096] Process B4: Calculate the loss function of the initial large model based on the prediction results.

[0097] It should be noted that the loss function uses the cross entropy loss function to measure the difference between the model prediction results and the true label.

[0098] Step B5: Determine whether the loss function meets the preset value. If the loss function meets the preset value, execute step B6; if the loss function does not meet the preset value, execute step B7.

[0099] Process B6: If the loss function meets the preset value, the initial large model is optimized using the test set to obtain the final large model.

[0100] It is understandable that through multiple iterative training, until the model's loss function converges, and on the test set, the model's accuracy, recall rate, F1 value and other indicators reach ideal results, the final large model is determined.

[0101] Process B7: If the loss function does not meet the preset value, the parameters of the initial large model are updated according to the adaptive moment estimation optimization algorithm, and the process returns to process B4.

[0102] It should be noted that the Adaptive Moment Estimation (Adam) optimization algorithm is used to minimize the loss function.

[0103] Step S103: If the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment, an energy scheduling strategy is generated and output based on the prediction result, the target electricity consumption curve data, and the target peak and valley electricity price data.

[0104] In the specific implementation of step S103, if the prediction result output by the large model indicates that the factory to be tested is equipped with photovoltaic equipment and energy storage equipment, the optimal energy scheduling strategy is generated and output based on the prediction result, target electricity consumption curve data and target peak and valley electricity price data.

[0105] It is understood that the energy scheduling strategy includes, but is not limited to, the power usage strategy and corresponding power distribution of the factory under test during different time periods. In other words, when to use photovoltaic power, when to use energy storage power, and when to purchase power from the grid.

[0106] For example, when photovoltaic power is sufficient, photovoltaic power will be used first; if photovoltaic power is insufficient or cannot meet demand, power will be supplemented by the energy storage system; when neither photovoltaic power nor energy storage power can meet demand, the factory under test will purchase electricity from the power grid to ensure stable power supply for production needs.

[0107] It should be noted that the specific process of generating the energy scheduling strategy is as follows (process C1 to process C4):

[0108] Process C1: Extract the photovoltaic installed capacity value and energy storage battery capacity value from the prediction results.

[0109] In the specific implementation process C1, the photovoltaic installed capacity value P is extracted from the prediction results output by the large model. pv (unit kW) and energy storage battery capacity value E es (unit: kWh), which is the predicted value of photovoltaic installed capacity and energy storage battery capacity.

[0110] Process C2: Obtaining grid power supply information and preset time period.

[0111] When specifically implementing process C2, grid power supply information (eg, maximum grid power supply) and a preset time period (eg, one month) are obtained from the grid.

[0112] Process C3: Generate an energy scheduling strategy based on grid power supply information, preset time period, photovoltaic installed capacity value, energy storage battery capacity value, target power consumption curve data and peak and valley electricity prices.

[0113] When specifically implementing process C3, first, an objective function is constructed based on the preset time period, peak and valley electricity prices, and target electricity consumption curve data.

[0114] The objective function is to minimize the factory's monthly electricity expenditure C, as shown in formula (1).

[0115] (1)

[0116] In formula (1), T is the total number of time periods in a month (assuming it is in hours, T=720). t is the electricity price in period t. u t The amount of electricity purchased from the grid by the factory under test during period t (unit: kWh).

[0117] Secondly, the constraints are determined based on the target electricity consumption curve data, photovoltaic installed capacity value, energy storage battery capacity value and grid power supply information.

[0118] It should be noted that the constraints are as shown in formulas (2) to (4).

[0119] First: The power demand constraint of the factory under test is shown in formula (2).

[0120] (2)

[0121] In formula (2), D t is the actual electricity demand of the factory under test during period t (unit: kWh). t is the power generation power of the photovoltaic equipment in time period t (unit: kW).

[0122] Assuming that the energy dispatch strategy can be generated based on weather forecast data and production plan (as shown in the following process D1 to process D5), in formula (2), d t is the discharge power of the energy storage device in time period t (unit: kW).

[0123] Second: The operating limitations of photovoltaic and energy storage equipment are shown in formula (3).

[0124] 0≤h t ≤P pv , 0≤d t ≤E es (3)

[0125] In formula (3), h t P is the power generation power of the photovoltaic equipment in period t (unit: kW). pvis the photovoltaic installed capacity value in the prediction result (unit: kW). t E is the discharge power of the energy storage device during time period t (unit: kW). es is the energy storage battery capacity value in the prediction result (unit: kWh).

[0126] Third: The power supply capacity constraint of the power grid is shown in formula (4).

[0127] u t ≤U max (4)

[0128] In formula (4), U max is the maximum power supply (kWh) that the grid can provide to the plant under test during time period t. t The amount of electricity purchased from the grid by the factory under test during period t (unit: kWh).

[0129] Finally, the energy scheduling strategy is generated based on the preset optimization algorithm, objective function and constraints.

[0130] The preset optimization algorithm may be a genetic algorithm, a particle swarm optimization algorithm or a linear programming algorithm.

[0131] For example, a genetic algorithm can be used to solve the problem. First, the amount of electricity purchased in each period u t The genes are encoded into a chromosome. Next, an initial population is randomly generated and continuously evolved through genetic operations such as selection, crossover, and mutation. Ultimately, after multiple iterations, the optimal chromosome is found, resulting in the optimal energy scheduling strategy.

[0132] In some application embodiments, the power consumption of the factory to be tested is monitored in real time, and the energy scheduling strategy is adjusted according to the power consumption.

[0133] Process C4: Output the energy scheduling strategy.

[0134] During the specific implementation of process C4, the energy scheduling strategy is output through the interactive interface.

[0135] It is understandable that the prediction results can also be output through an interactive interface (developed based on a Web framework (Django)) and the user's adjustment and optimization of the energy scheduling strategy can be received through the interactive interface.

[0136] In actual applications, energy-saving suggestion information or electricity usage reports can also be generated to help users manage energy.

[0137] In some specific embodiments, robustness analysis can be performed on the generated energy scheduling strategy to account for the impact of weather changes or other uncertainties on photovoltaic power generation, as well as the uncertainty of factory production plans. By simulating different weather conditions and production plan changes, the stability and effectiveness of the strategy can be evaluated, effectively reducing electricity costs.

[0138] If it is found that the solution performs poorly in certain cases, the model can be adjusted and optimized, such as adding constraints or adjusting the weight of the objective function, to improve the robustness of the solution.

[0139] For example, the following processes D1 to D5 illustrate the process of generating an energy scheduling strategy by combining weather forecast data and production plans.

[0140] Process D1: Extract the photovoltaic installed capacity value and energy storage battery capacity value from the prediction results.

[0141] Process D2: Obtaining grid power supply information and preset time period.

[0142] Process D3: Obtain weather forecast data and production plan.

[0143] Process D4: Generate an energy scheduling strategy based on weather forecast data, production plan, grid power supply information, preset time period, photovoltaic installed capacity value, energy storage battery capacity value, target power consumption curve data and peak and valley electricity prices.

[0144] Process D5: Output the energy scheduling strategy.

[0145] It should be noted that the specific implementation principles of process D1 to process D5 are consistent with the implementation principles of the above-mentioned process C1 to process C4, and will not be repeated here.

[0146] Understandably, see Figure 2 , shows an example diagram of the electricity cost comparison of the factory to be tested provided by an embodiment of the present invention. Figure 2 The daily electricity costs corresponding to the energy scheduling strategy (curve 1) and the daily electricity costs corresponding to the traditional analysis method (curve 2) are shown in Figure 1. It is clear that, for most of the time, the energy scheduling strategy generated by the embodiment of the present invention is more effective in reducing the factory's daily electricity costs than the traditional analysis method. This indicates that the energy scheduling strategy generated by the embodiment of the present invention significantly reduces electricity costs.

[0147] In an embodiment of the present invention, by utilizing the powerful data analysis and prediction capabilities of the large model, it is possible to accurately infer whether the factory has been equipped with photovoltaic equipment and energy storage equipment, and analyze the relevant equipment parameters, providing an important basis for energy management. Based on these prediction results, peak and valley electricity price data, electricity consumption curve data and other information, the optimal energy scheduling strategy is generated for the factory to be tested, which helps to reasonably arrange electricity consumption, reduce electricity costs, and improve energy utilization efficiency. This method manages factory energy intelligently, automatically analyzes real-time data and makes decisions, reduces manual intervention, improves management efficiency and accuracy, and thus helps the factory achieve energy conservation, emission reduction and sustainable development. In addition, the large model has strong versatility and adaptability, can adapt to the electricity consumption characteristics and energy management needs of different factories, and flexibly adjust and optimize according to actual conditions to meet the energy management requirements of different factories in different time periods.

[0148] Corresponding to the method for generating energy scheduling strategy based on photovoltaic energy storage equipment provided in the above embodiment of the present invention, see Figure 3 , shows a structural block diagram of a device for generating an energy scheduling strategy based on photovoltaic energy storage equipment provided by an embodiment of the present invention.

[0149] It should be noted that the device is developed using Python and a related deep learning framework (such as PyTorch). The device includes: a target data collection unit 301, a large model prediction unit 302, and a generation unit 303.

[0150] It is understandable that the target data collection unit 301 is connected to the data interface of the power monitoring system and the data interface of the electric meter equipment of the factory to be tested; the target data collection unit 301 is also connected to the data interface of the power grid.

[0151] The target data collection unit 301 is used to collect peak-valley electricity price data and power consumption curve data of the factory to be tested and perform preprocessing to obtain target power consumption curve data and target peak-valley electricity price data.

[0152] It should be noted that the target data collection unit 301 uses a Python data analysis library (such as Pandas, Numpy) for preprocessing.

[0153] The large model prediction unit 302 is used to input the target electricity consumption curve data and the target peak and valley electricity price data into the large model for prediction to obtain a prediction result; the large model is pre-trained based on the historical electricity consumption curve data.

[0154] The generating unit 303 is configured to generate and output an energy scheduling strategy based on the prediction result, the target power consumption curve data, and the target peak-valley electricity price data if the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment.

[0155] In an embodiment of the present invention, by utilizing the powerful data analysis and prediction capabilities of the large model, it is possible to accurately infer whether the factory has been equipped with photovoltaic equipment and energy storage equipment, and analyze the relevant equipment parameters, providing an important basis for energy management. Based on these prediction results, peak and valley electricity price data, electricity consumption curve data and other information, the optimal energy scheduling strategy is generated for the factory to be tested, which helps to reasonably arrange electricity consumption, reduce electricity costs, and improve energy utilization efficiency. By intelligently managing factory energy, automatically analyzing real-time data and making decisions, reducing manual intervention, and improving management efficiency and accuracy, the factory can achieve energy conservation, emission reduction and sustainable development. In addition, the large model has strong versatility and adaptability, can adapt to the electricity consumption characteristics and energy management needs of different factories, and flexibly adjust and optimize according to actual conditions to meet the energy management requirements of different factories in different time periods.

[0156] Combine Figure 3 As shown in the content, the device also includes: a sample data collection unit, a division unit, a construction unit, a calculation unit, a judgment unit, an optimization unit and an update unit.

[0157] The sample data collection unit is used to collect historical peak and valley electricity price data and historical electricity consumption curve data, and pre-process the historical peak and valley electricity price data and historical electricity consumption curve data to obtain a sample data set.

[0158] The partitioning unit is used to divide the sample data set into a training set and a test set.

[0159] The construction unit is used to construct an initial large model and input the training set into the initial large model to obtain the prediction results corresponding to the training set.

[0160] The calculation unit is used to calculate the loss function of the initial large model based on the prediction results.

[0161] The judgment unit is used to judge whether the loss function meets the preset value.

[0162] The optimization unit is used to optimize the initial large model using the test set to obtain the final large model if the loss function meets the preset value.

[0163] The updating unit is used to update the parameters of the initial large model according to the adaptive moment estimation optimization algorithm if the loss function does not meet the preset value, and return to the execution calculation unit.

[0164] Combine Figure 3 As shown in the content, the target data collection unit 301 includes: a first collection module, a second collection module and a pre-processing module.

[0165] The first collection module is used to collect peak and valley electricity price data within a preset time period.

[0166] The second collection module is used to collect power and electricity consumption information within a preset time period from the power monitoring system and electricity meter equipment of the factory to be tested, and obtain electricity consumption curve data.

[0167] The preprocessing module is used to preprocess the peak-valley electricity price data and the electricity consumption curve data to obtain the target electricity consumption curve data and the target peak-valley electricity price data.

[0168] Combine Figure 3 The content shown, the generating unit 303, includes: an extracting module, an acquiring module, a generating module and an output module.

[0169] The extraction module is used to extract the photovoltaic installed capacity value and the energy storage battery capacity value from the prediction results.

[0170] The acquisition module is used to obtain power grid power supply information and preset time periods.

[0171] The generation module is used to generate an energy scheduling strategy based on grid power supply information, preset time periods, photovoltaic installed capacity values, energy storage battery capacity values, target power consumption curve data, and peak and valley electricity prices.

[0172] The output module is used to output the energy scheduling strategy.

[0173] Combine Figure 3 The content shown, generating module, includes: constructing submodule, determining submodule and generating submodule.

[0174] The construction submodule is used to construct the objective function according to the preset time period, peak and valley electricity prices and target electricity consumption curve data.

[0175] The determination submodule is used to determine the constraint conditions based on the target power consumption curve data, photovoltaic installed capacity value, energy storage battery capacity value and grid power supply information.

[0176] The generation submodule is used to generate energy scheduling strategies based on preset optimization algorithms, objective functions and constraints.

[0177] Combine Figure 3 As shown in the content, the device also includes: an acquisition unit, a strategy generation unit and an output unit.

[0178] The acquisition unit is used to obtain weather forecast data and production plans.

[0179] The strategy generation unit is used to generate energy scheduling strategies based on weather forecast data, production plans, grid power supply information, preset time periods, photovoltaic installed capacity values, energy storage battery capacity values, target power consumption curve data and peak and valley electricity prices.

[0180] The output unit is used to output the energy scheduling strategy.

[0181] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple. For relevant parts, refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0182] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0183] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating an energy scheduling strategy based on photovoltaic energy storage equipment, characterized in that: The method comprises: Collect peak and valley electricity price data and the power consumption curve data of the factory to be tested and pre-process them to obtain target power consumption curve data and target peak and valley electricity price data; Inputting the target electricity consumption curve data and the target peak-valley electricity price data into a large model for prediction to obtain a prediction result; the large model is pre-trained based on historical electricity consumption curve data; If the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment, an energy scheduling strategy is generated and output based on the prediction result, the target electricity consumption curve data and the target peak-valley electricity price data.

2. The method according to claim 1, characterized in that The process of pre-training a large model based on historical electricity consumption curve data includes: Collecting historical peak-valley electricity price data and historical electricity consumption curve data, and preprocessing the historical peak-valley electricity price data and the historical electricity consumption curve data to obtain a sample data set; Dividing the sample data set into a training set and a test set; Constructing an initial large model, and inputting the training set into the initial large model to obtain a prediction result corresponding to the training set; Calculating the loss function of the initial large model according to the prediction result; Determine whether the loss function meets a preset value; If the loss function satisfies a preset value, the initial large model is optimized using the test set to obtain a final large model; If the loss function does not meet the preset value, the parameters of the initial large model are updated according to the adaptive moment estimation optimization algorithm, and the step of calculating the loss function of the initial large model according to the prediction result is returned to be executed.

3. The method according to claim 1, characterized in that The collecting of peak and valley electricity price data and electricity consumption curve data of the tested factory and preprocessing to obtain target electricity consumption curve data and target peak and valley electricity price data includes: Collect peak and valley electricity price data within a preset time period; Collect power and electricity consumption information within a preset time period from the power monitoring system and electricity meter equipment of the factory to be tested, and obtain electricity consumption curve data; The peak-valley electricity price data and the electricity consumption curve data are preprocessed to obtain target electricity consumption curve data and target peak-valley electricity price data.

4. The method according to claim 1, wherein The generating and outputting of an energy scheduling strategy based on the prediction result, the target electricity consumption curve data and the peak-valley electricity price includes: Extracting the photovoltaic installed capacity value and the energy storage battery capacity value from the prediction results; Obtain grid power supply information and preset time periods; Generate an energy scheduling strategy based on the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data and the peak and valley electricity price; The energy scheduling strategy is output.

5. The method according to claim 4, characterized in that The generating of an energy scheduling strategy based on the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data and the peak-valley electricity price includes: Constructing an objective function according to the preset time period, the peak-valley electricity price and the target electricity consumption curve data; Determining constraint conditions based on the target power consumption curve data, the photovoltaic installed capacity value, the energy storage battery capacity value, and the grid power supply information; An energy scheduling strategy is generated according to a preset optimization algorithm, the objective function and the constraint conditions.

6. The method according to claim 4, characterized in that After obtaining the grid power supply information and the preset time period, it also includes: Obtain weather forecast data and production plans; Generate an energy scheduling strategy based on the weather forecast data, the production plan, the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data and the peak and valley electricity price; The energy scheduling strategy is output.

7. A device for generating an energy scheduling strategy based on photovoltaic energy storage equipment, characterized in that: The device comprises: A target data collection unit is used to collect peak and valley electricity price data and electricity consumption curve data of the factory to be tested and pre-process them to obtain target electricity consumption curve data and target peak and valley electricity price data; A large model prediction unit is used to input the target electricity consumption curve data and the target peak and valley electricity price data into the large model for prediction to obtain a prediction result; the large model is pre-trained based on historical electricity consumption curve data; A generating unit is configured to generate and output an energy scheduling strategy based on the prediction result, the target electricity consumption curve data, and the target peak-valley electricity price data if the prediction result indicates that the factory to be tested is equipped with photovoltaic energy storage equipment.

8. The device according to claim 7, characterized in that The device further comprises: a sample data collection unit, configured to collect historical peak-valley electricity price data and historical electricity consumption curve data, and pre-process the historical peak-valley electricity price data and the historical electricity consumption curve data to obtain a sample data set; A division unit, configured to divide the sample data set into a training set and a test set; A construction unit, configured to construct an initial large model, and input the training set into the initial large model to obtain a prediction result corresponding to the training set; A calculation unit, configured to calculate a loss function of the initial large model according to the prediction result; A judgment unit, configured to judge whether the loss function satisfies a preset value; an optimization unit, configured to optimize the initial large model using the test set to obtain a final large model if the loss function satisfies a preset value; An updating unit is used to update the parameters of the initial large model according to an adaptive moment estimation optimization algorithm if the loss function does not meet a preset value, and return to execute the calculation unit.

9. The device according to claim 7, characterized in that The target data collection unit includes: The first collection module is used to collect peak and valley electricity price data within a preset time period; The second collection module is used to collect power and electricity consumption information within a preset time period from the power monitoring system and electricity meter equipment of the factory to be tested, and obtain electricity consumption curve data; The preprocessing module is used to preprocess the peak-valley electricity price data and the electricity consumption curve data to obtain target electricity consumption curve data and target peak-valley electricity price data.

10. The device according to claim 7, characterized in that The generating unit comprises: An extraction module, configured to extract the photovoltaic installed capacity value and the energy storage battery capacity value from the prediction result; An acquisition module is used to obtain power grid power supply information and preset time periods; A generation module, configured to generate an energy scheduling strategy based on the grid power supply information, the preset time period, the photovoltaic installed capacity value, the energy storage battery capacity value, the target power consumption curve data, and the peak and valley electricity price; The output module is used to output the energy scheduling strategy.

Images