EMS energy management method and system based on photovoltaic power station

Abstract

The invention relates to an EMS energy management method and system based on a photovoltaic power station. The method comprises the following steps: S1, acquiring an initial strategy based on a target power station, and setting an actual working process of the target power station by taking the initial strategy as a main strategy; s2, acquiring a corresponding shadow strategy from a preset strategy library based on the characteristic parameters of the target power station, and constructing a virtual working process corresponding to the target power station by using the shadow strategy; s3, predicting the generating capacity of the target power station in the target time period according to the actual working process to obtain a first prediction result, and predicting the generating capacity of the target power station in the target time period according to the virtual working process to obtain a second preset result; and S4, obtaining the actual generating capacity of the target power station corresponding to the target time period, comparing the actual generating capacity with the first prediction result and the second prediction result, and switching the target power station to take the shadow strategy as the main strategy when the second preset result is close to the actual generating capacity. According to the invention, strategy optimization of the power station can be rapidly realized.

Description

Technical Field

[0001] This invention relates to the field of energy management technology, and more specifically, to an EMS energy management method and system based on photovoltaic power plants. Background Technology

[0002] Current energy management systems (EMS) typically employ static rules or fixed models for strategy selection and adjustment during energy management. This makes it difficult for the entire strategy to adapt to dynamic energy demands such as fluctuations in renewable energy and changes in load. Each optimization requires starting from scratch, wasting computing resources. Furthermore, the entire process relies on manual experience for adjustments, resulting in slow response times and an inability to achieve real-time optimal scheduling. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an EMS energy management method and system based on photovoltaic power plants, addressing the aforementioned technical deficiencies of the prior art.

[0004] The technical solution adopted by this invention to solve its technical problem is: to construct an EMS energy management method based on a photovoltaic power station, the method comprising the following steps: S1. Obtain an initial strategy based on the target power station, and set the actual working process of the target power station using the initial strategy as the main strategy; S2. Obtain the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station, and construct the virtual working process corresponding to the target power station using the shadow strategy; S3. Based on the actual working process, predict the power generation of the target power station within the target time period to obtain a first prediction result; based on the virtual working process, predict the power generation of the target power station within the target time period to obtain a second preset result. S4. Obtain the actual power generation of the target power station corresponding to the target time period, compare the actual power generation with the first prediction result and the second prediction result, and when the second preset result is close to the actual power generation, switch the target power station to use the shadow strategy as the main strategy.

[0005] Preferably, in the EMS energy management method of the present invention, the method further includes: obtaining characteristic parameters corresponding to a known power station and weight values ​​corresponding to the characteristic parameters; constructing a strategy identifier corresponding to the known power station based on the characteristic parameters and the weight values; and generating the preset strategy library based on the strategy identifier and the strategy corresponding to the known power station. In step S2, obtaining the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station includes: The characteristic parameters of the target power station are obtained and matched with all policy identifiers in the preset policy library to obtain the policy corresponding to the policy identifier that matches the target power station as the shadow policy of the target power station.

[0006] Preferably, in the EMS energy management method of the present invention, the characteristic parameters corresponding to the power station include: the scene type of the power station, climate characteristics, equipment compatibility, data coverage and performance indicators.

[0007] Preferably, in the EMS energy management method of the present invention, the step of obtaining the characteristic parameters of the target power station and matching them with all policy identifiers in the preset policy library to obtain the policy corresponding to the policy identifier that matches the target power station as the shadow policy of the target power station includes: Obtain key parameters from the characteristic parameters of the target power station, and perform initial strategy selection based on the strategy identifier according to the key parameters; Based on the initial strategy selection results, the similarity between the feature parameters of the target power station and all strategy identifiers is obtained, as well as the sum of the products of the similarity and weight values ​​of all feature parameters of the strategy identifier. Based on the magnitude of the sum of the products, the strategy that matches the target power station among the strategies corresponding to the strategy identifier is obtained.

[0008] Preferably, in the EMS energy management method of the present invention, the method further includes: S5. Adjust the main strategy based on the target power station to obtain the final target strategy, and update the preset strategy library based on the target power station and the target strategy.

[0009] Preferably, in the EMS energy management method of the present invention, the step of adjusting the main strategy based on the target power station to obtain the final target strategy, and updating the preset strategy library based on the target power station and the target strategy, includes: Adjust the weight values ​​corresponding to the feature parameters according to the actual operating parameter requirements of the target power plant to update the strategy identifier corresponding to the main strategy; and / or The main strategy is optimized based on the actual working process of the target power plant to obtain the target strategy.

[0010] Preferably, in the EMS energy management method of the present invention, optimizing the main strategy based on the actual working process of the target power plant to obtain the target strategy includes: The power generation of the target power station is predicted based on the current weather parameters corresponding to the target power station to obtain the corresponding prediction result. The corresponding parameters of the main strategy are optimized according to the difference between the prediction result and the actual power generation. When the difference between the prediction result and the actual power generation of the target power station is less than a threshold value, the optimized main strategy is adopted as the target strategy.

[0011] Preferably, in the EMS energy management method of the present invention, in step S2, obtaining the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station includes: obtaining one or more strategies as corresponding shadow strategies based on the preset strategy library.

[0012] Preferably, in the EMS energy management method of the present invention, the method further includes: after completing step S4, continuing to execute step S2.

[0013] The present invention also provides an EMS energy management system based on a photovoltaic power plant, including a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program to implement the method described above.

[0014] The EMS energy management method and system based on photovoltaic power plants according to the present invention has the following beneficial effects: it can quickly realize the strategy optimization of the power plant. Attached Figure Description

[0015] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart of an embodiment of an EMS energy management method for a photovoltaic power station according to the present invention; Figure 2 This is a flowchart of another embodiment of the EMS energy management method for a photovoltaic power station according to the present invention; Figure 3 This is a flowchart of another embodiment of the EMS energy management method for a photovoltaic power station according to the present invention. Detailed Implementation

[0016] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0017] like Figure 1 The figure shows an embodiment of an EMS energy management method based on a photovoltaic power plant according to the present invention. Figure 1An embodiment of the EMS energy management method based on a photovoltaic power station of the present invention is shown, including the following steps: S1, obtaining an initial strategy based on the target power station, and setting the actual working process of the target power station with the initial strategy as the main strategy; S2, obtaining the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station, and constructing a virtual working process corresponding to the target power station with the shadow strategy; S3, predicting the power generation of the target power station in the target time period according to the actual working process to obtain a first prediction result, and predicting the power generation of the target power station in the target time period according to the virtual working process to obtain a second preset result; S4, obtaining the actual power generation of the target power station corresponding to the target time period, comparing the actual power generation with the first prediction result and the second prediction result, and when the second preset result is close to the actual power generation, switching the target power station to use the shadow strategy as the main strategy.

[0018] Based on step S1, during the photovoltaic power plant design process, an initial strategy can be obtained based on the target power plant when it starts operating or is already running. This initial strategy can be an existing operating strategy of the target power plant or a strategy derived from existing power plants. The target power plant is then configured to operate using this initial strategy as its primary strategy. Simply put, the actual operating process of the target power plant at this point is determined by the set primary strategy. Furthermore, it can be understood that the current primary strategy of the target power plant may not be its optimal operating strategy.

[0019] Based on step S2, a shadow strategy corresponding to the target power station is selected according to its characteristic parameters. This shadow strategy is used to construct a virtual operating process simulating the target power station's operation. That is, when the target power station operates based on the main strategy, its corresponding operation is simulated using the shadow strategy. However, the preset strategy library contains multiple existing strategies, not all of which are applicable to the target power station. Therefore, it is necessary to analyze the characteristic parameters of the target power station to obtain a matching shadow strategy.

[0020] In one embodiment, based on step S2 above, a corresponding shadow strategy is obtained from a preset strategy library based on the characteristic parameters of the target power station; this includes: obtaining one or more strategies from the preset strategy library as corresponding shadow strategies. It can be understood that when selecting a shadow strategy corresponding to the target power station, one or more strategies can be selected as the shadow strategy for that target power station. When multiple shadow strategies are obtained, a virtual working process is constructed for each shadow strategy, that is, the shadow strategy corresponds one-to-one with the constructed virtual working process of the target power station.

[0021] Based on step S3 and the above process, it can be understood that the target power station at this point can be considered to have two or more strategies operating, each predicting the power generation of the target power station based on its operating process. For example, the power generation prediction result for a certain time period is obtained based on the actual operating process of the target power station, which is the first prediction result; the power generation prediction result for that time period is obtained based on the virtual operating process of the target power station, which is the second prediction result. When the target power station corresponds to multiple virtual operating processes, it can be understood that multiple second prediction results can be obtained. For example, predicting the photovoltaic power generation for the next 24 hours on an hourly basis. During the operation of the power station, the power generation for future time periods is usually predicted to determine how much electricity should be stored in the energy storage system, thereby formulating strategies to achieve economic optimization while ensuring load demand.

[0022] Based on step S4, after obtaining the above prediction results, when the target power station completes its actual operation for the target time period, the actual power generation of the target power station is obtained, and the actual power generation is compared with the first prediction result and the second prediction result. When the first prediction result is close to the actual power generation, it indicates that the current strategy of the target power station is more suitable for the target power station, and the current strategy of the target power station continues to be the main strategy. When the second prediction result is close to the actual power generation, it indicates that the shadow strategy corresponding to the target power station is more suitable for the target power station. At this time, it is necessary to switch the main strategy of the target power station and use the corresponding shadow strategy as the main strategy of the target power station. It should be understood that among multiple second prediction results, the shadow strategy corresponding to the second prediction result that is closer to the actual discharge is selected as the main strategy of the target power station.

[0023] Based on the above process, multiple strategies can be implemented in parallel, which can greatly reduce the optimization process of the target power plant's working strategy.

[0024] In one embodiment, the EMS energy management method based on photovoltaic power plants of the present invention further includes: obtaining characteristic parameters and weight values ​​corresponding to known power plants, constructing strategy identifiers corresponding to known power plants based on the characteristic parameters and weight values, and generating a preset strategy library based on the strategy identifiers and strategies corresponding to known power plants; in step S2, obtaining the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power plant includes: obtaining the characteristic parameters of the target power plant and matching them with all strategy identifiers in the preset strategy library to obtain the strategy corresponding to the strategy identifier that matches the target power plant as the shadow strategy of the target power plant.

[0025] Specifically, a pre-defined strategy library can be constructed based on known power plants. For several optimized known power plants, characteristic parameters and their corresponding weight values ​​are obtained. For each known power plant, a strategy identifier is constructed based on its characteristic parameters and weight values. A correspondence is established between the strategy (i.e., the main strategy) and its corresponding strategy identifier for each known power plant. Based on this correspondence, strategies and strategy identifiers are stored, thus obtaining the pre-defined strategy library. In the process of obtaining shadow strategies from the pre-defined strategy library based on the characteristic parameters of the target power plant, the characteristic parameters of the target power plant can be obtained first. Then, the characteristic parameters of the target power plant are matched with the strategy identifiers corresponding to each strategy in the pre-defined strategy library to obtain strategy identifiers that meet the matching requirements. The strategy corresponding to this strategy identifier is then used as the shadow strategy for the target power plant.

[0026] In a specific implementation, the characteristic parameters corresponding to the power station include the power station's scenario type, climate characteristics, equipment compatibility, data coverage, and performance indicators. Specifically, the characteristic parameters are categorized based on their impact on the power station's performance. For example, the scenario type of the power station can be set as one of the following: desert power station, high-latitude power station, distributed rooftop, floating marine, agricultural photovoltaic (bifacial modules), wind farm, microgrid, etc. The climate characteristics of the power station can be set as one of the following: arid, rainy, high-irradiance, etc. Equipment compatibility can be set, for example, the power station can be compatible with one or more of the following: inverter model A, photovoltaic panel type B, etc. Data coverage can be set, for example, the irradiance of the power station can be set to 300-1500 W / m², etc. Performance indicators can be set, for example, the performance of the power station can be set to meet the requirement of root mean square error ≤ 4 kW, where the root mean square error measures the difference between the predicted value and the actual value; the smaller the value, the more accurate the prediction.

[0027] In one specific embodiment, weight values ​​are set for the above feature parameters respectively. For example, the weight value of scene type is set to 0.3, the weight value of climate feature is set to 0.25, the weight value of device compatibility is set to 0.2, the weight value of data coverage is set to 0.15, and the weight value of performance index is set to 0.1.

[0028] In one embodiment, the process of obtaining feature parameters of a target power station and matching them with all policy identifiers in a preset policy library to obtain the policy corresponding to the policy identifier that matches the target power station as the shadow policy of the target power station includes: obtaining key parameters from the feature parameters of the target power station; performing initial policy selection based on the policy identifier according to the key parameters; obtaining the similarity between the feature parameters of the target power station and all policy identifiers based on the initial policy selection results, as well as the sum of the products of the similarity and weight values ​​of all feature parameters of the policy identifier; and obtaining the policy that matches the target power station from the policies corresponding to the policy identifier according to the magnitude of the sum of the products.

[0029] Specifically, when performing policy matching based on policy identifiers, an initial policy selection is first performed based on the key parameters of the target power station. For example, policies that do not meet the requirements are eliminated based on scenario type and equipment incompatibility. Then, a weighted score is calculated based on the remaining policies. The process of obtaining the weighted score is as follows: in, The weight of the i-th feature parameter, Let be the similarity of the i-th feature parameter, with a value ranging from 0 to 1.

[0030] The similarity calculation process is as follows: for the data coverage range, it is the ratio of the overlap between the new range and the strategy training range to the new range; for the performance index, it is the ratio of the absolute difference between the new item value and the strategy nominal value to the strategy nominal value.

[0031] A specific embodiment is used for illustration, matching is performed based on strategy A and strategy B in the preset strategy library. For strategy A, the device model does not match and is directly eliminated. For strategy B, the scene type matching Sim=1, the weight is 0.3X1=0.3, the climate feature matching Sim=0.9, the weight score is 0.225, the device model matching Sim=1, the weight score is 0.2, the irradiance range overlap is 80%, the weight score is 0.15×0.8=0.12, RMSE=6kW, based on the new project requirement of ≤8kW, Sim=1-|6-8| / 8=0.75, the weight score is 0.1X0.75=0.075. The final total weighted score is 0.3+0.225+0.2+0.12+0.075=0.92. When there is no strategy with a higher score, strategy B is selected as the shadow strategy.

[0032] like Figure 2As shown, in one embodiment of the EMS energy management method of the present invention, it further includes: S5, adjusting the main strategy based on the target power station to obtain the final target strategy, and updating the preset strategy library based on the target power station and the target strategy. Specifically, after obtaining the main strategy of the target power station based on step S4, the main strategy can also be adjusted based on the main strategy operation results. In a specific embodiment, adjusting the main strategy based on the target power station to obtain the final target strategy, and updating the preset strategy library based on the target power station and the target strategy, includes: optimizing the main strategy based on the actual working process of the target power station to obtain the target strategy. For example, calculation and testing are performed based on the selected strategy B, such as injecting real-time data from the target power station and running the strategy for 24 hours. RMSE, response time, and other measured performance scores are calculated. If the measured score is lower than a threshold (e.g., RMSE > 10kW), the strategy hyperparameters such as the neural network learning rate are adjusted. Transfer learning is then performed using the data from the target power station to train the model and obtain the final target record.

[0033] In another embodiment, the main strategy is adjusted based on the target power plant to obtain the final target strategy, and the preset strategy library is updated based on the target power plant and the target strategy. This includes: adjusting the weight values ​​corresponding to the feature parameters according to the actual operating parameter requirements of the target power plant to update the strategy identifier corresponding to the main strategy. That is, based on the actual operating requirements of the target power plant, for example, if the target power plant is sensitive to response speed, the weight of performance indicators is increased, for example, from 0.1 to 0.3, while the data coverage requirement is low, for example, if historical data has high noise, the weight of data range is reduced, for example, from 0.15 to 0.05. The updated main strategy obtained based on the optimized weights is the target strategy.

[0034] During policy-based model training, the training process can be terminated based on a set maximum number of iterations. Based on the adjusted target policy, the target power station can be treated as a known power station to optimize the pre-defined policy library. For details, refer to the process of establishing the pre-defined policy library above. This involves storing the target power station's policy identifier and its target policy in the pre-defined policy library.

[0035] In one embodiment, the main strategy is optimized based on the actual operating process of the target power plant to obtain the target strategy. This includes: predicting the power generation of the target power plant based on the current weather parameters corresponding to the target power plant to obtain the corresponding prediction result; optimizing the corresponding parameters of the main strategy based on the difference between the prediction result and the actual power generation; and adopting the optimized main strategy as the target strategy when the difference between the prediction result and the actual power generation of the target power plant is less than a threshold value. For example, the photovoltaic power generation power of the next hour is predicted based on weather data (such as solar irradiance and temperature). The solar irradiance parameters of the target power plant are obtained based on the main strategy, in W / m², ambient temperature, and photovoltaic panel temperature, in °C. The power generation power is predicted based on the main strategy, for example, the model predicts that the power generation power of the next hour will be 66 kW. After one hour, the actual power generation power of the target power plant is measured to be 70 kW, with an error of 4 kW from the predicted value of 66 kW. The main strategy is adjusted, and a prediction is made again to make the prediction result as close as possible to the actual measurement result.

[0036] like Figure 3 As shown, in one embodiment of the EMS energy management method of the present invention, the method further includes continuing to execute step S2 after completing step S4. That is, after switching to obtain the main strategy of the target power station, optimization can continue based on the current main strategy to match it according to a preset strategy library to confirm whether a better main strategy can be obtained. Step S2 can be triggered as needed, for example, periodically, or based on changes in the parameters of the target power station, to ensure that the target power station is in the optimal strategy as much as possible.

[0037] Furthermore, the present invention provides an EMS energy management system based on a photovoltaic power plant, which has the function of implementing the corresponding steps in the above-described method. Each function can be implemented by hardware or by hardware executing corresponding software. The corresponding hardware or software includes one or more modules corresponding to the above functions. That is, the steps in the above-described method are executed by one or more modules respectively. The specific cooperation between the modules can be referred to the specific process of the above-described method, and will not be repeated here.

[0038] Furthermore, the EMS energy management system based on a photovoltaic power station according to the present invention may further include a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the above-described method. Specifically, according to embodiments of the present invention, the processes described above with reference to the flowchart can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, when the computer program is downloaded, installed, and executed by an electronic device, it performs the functions defined in the methods of the embodiments of the present invention. The electronic device in the present invention may be a terminal such as a laptop, desktop computer, tablet computer, or smartphone, or it may be a server.

[0039] It is understood that the above embodiments only illustrate preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can freely combine the above technical features without departing from the concept of the present invention, and can also make several modifications and improvements, all of which fall within the protection scope of the present invention. Therefore, all equivalent transformations and modifications made with respect to the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. An EMS energy management method based on photovoltaic power plants, characterized in that, The method includes the following steps: S1. Obtain an initial strategy based on the target power station, and set the actual working process of the target power station using the initial strategy as the main strategy; S2. Obtain the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station, and construct the virtual working process corresponding to the target power station using the shadow strategy; S3. Based on the actual working process, predict the power generation of the target power station within the target time period to obtain a first prediction result; based on the virtual working process, predict the power generation of the target power station within the target time period to obtain a second preset result. S4. Obtain the actual power generation of the target power station corresponding to the target time period, compare the actual power generation with the first prediction result and the second prediction result, and when the second preset result is close to the actual power generation, switch the target power station to use the shadow strategy as the main strategy.

2. The EMS energy management method according to claim 1, characterized in that, The method further includes: obtaining feature parameters corresponding to a known power station and weight values ​​corresponding to the feature parameters; constructing a strategy identifier corresponding to the known power station based on the feature parameters and the weight values; and generating the preset strategy library based on the strategy identifier and the strategy corresponding to the known power station. In step S2, obtaining the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station includes: The characteristic parameters of the target power station are obtained and matched with all policy identifiers in the preset policy library to obtain the policy corresponding to the policy identifier that matches the target power station as the shadow policy of the target power station.

3. The EMS energy management method according to claim 2, characterized in that, The characteristic parameters corresponding to the power station include: the power station's scene type, climate characteristics, equipment compatibility, data coverage, and performance indicators.

4. The EMS energy management method according to claim 2, characterized in that, The step of obtaining the characteristic parameters of the target power station and matching them with all policy identifiers in the preset policy library to obtain the policy corresponding to the policy identifier that matches the target power station as the shadow policy of the target power station includes: Obtain key parameters from the characteristic parameters of the target power station, and perform initial strategy selection based on the strategy identifier according to the key parameters; Based on the initial strategy selection results, the similarity between the feature parameters of the target power station and all strategy identifiers is obtained, as well as the sum of the products of the similarity and weight values ​​of all feature parameters of the strategy identifier. Based on the magnitude of the sum of the products, the strategy that matches the target power station among the strategies corresponding to the strategy identifier is obtained.

5. The EMS energy management method according to claim 2, characterized in that, The method further includes: S5. Adjust the main strategy based on the target power station to obtain the final target strategy, and update the preset strategy library based on the target power station and the target strategy.

6. The EMS energy management method according to claim 5, characterized in that, The step of adjusting the main strategy based on the target power station to obtain the final target strategy, and updating the preset strategy library based on the target power station and the target strategy, includes: Adjust the weight values ​​corresponding to the feature parameters according to the actual operating parameter requirements of the target power plant to update the strategy identifier corresponding to the main strategy; and / or The main strategy is optimized based on the actual working process of the target power plant to obtain the target strategy.

7. The EMS energy management method according to claim 6, characterized in that, The optimization of the main strategy based on the actual working process of the target power plant to obtain the target strategy includes: The power generation of the target power station is predicted based on the current weather parameters corresponding to the target power station to obtain the corresponding prediction result. The corresponding parameters of the main strategy are optimized according to the difference between the prediction result and the actual power generation. When the difference between the prediction result and the actual power generation of the target power station is less than a threshold value, the optimized main strategy is adopted as the target strategy.

8. The EMS energy management method according to claim 1, characterized in that, In step S2, obtaining the corresponding shadow strategy from the preset strategy library based on the characteristic parameters of the target power station includes: obtaining one or more strategies as corresponding shadow strategies based on the preset strategy library.

9. The EMS energy management method according to claim 1, characterized in that, The method further includes: continuing to execute step S2 after completing step S4.

10. An EMS energy management system based on a photovoltaic power station, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is used to execute the computer program to implement the method as described in any one of claims 1 to 9.

Images