Dynamic cooperative control method for photovoltaic wind power complementary power generation system and energy storage
By constructing a hierarchical prediction model and a dynamic collaborative control method that monitors deviations and compensates for them in real time, the shortcomings of data acquisition and scheduling control in photovoltaic-wind power complementary power generation systems are solved, achieving precise power output matching and grid stability assurance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-01-15
- Publication Date
- 2026-04-10
AI Technical Summary
In existing photovoltaic-wind power complementary power generation systems, the lack of targeted dimensional constraints and time calibration in multi-source data acquisition leads to insufficient input accuracy of prediction models. Furthermore, the dispatch control does not take into account the equipment's adjustment capabilities and grid security constraints, resulting in problems such as curtailment of solar and wind power, overcharging and over-discharging of energy storage, and fluctuations in grid voltage and frequency.
By collecting data on new energy output, environment, and load, a hierarchical prediction model is constructed. Combined with the energy storage status, dynamic and coordinated control is carried out to monitor and compensate for output deviations in real time, ensuring accurate data matching and equipment regulation capabilities in accordance with grid safety constraints.
It significantly improved the renewable energy absorption rate and system stability, reduced power output prediction errors, avoided overcharging and over-discharging of energy storage, and ensured grid stability and equipment safety.
Smart Images
Figure CN121840792A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of new energy power generation control technology, and more specifically, to a dynamic coordinated control method for photovoltaic and wind power complementary power generation systems and energy storage. Background Technology
[0002] As core forms of renewable energy utilization, photovoltaic and wind power occupy an important position in the energy structure transformation due to their clean and low-carbon advantages. However, their inherent intermittency and volatility have always constrained the stable operation of the system. Photovoltaic output is easily affected by cloud cover and day-night cycles, while wind power output depends on wind speed changes and is subject to gusts. This makes it difficult to accurately match the output of new energy sources with load demand and grid acceptance capacity.
[0003] Existing photovoltaic-wind power complementary power generation systems mostly adopt fixed dispatch strategies. Multi-source data acquisition lacks targeted dimensional constraints and time calibration, resulting in prominent issues of data redundancy and time misalignment, leading to insufficient input accuracy of subsequent prediction models. Output prediction does not fully adapt to the differentiated characteristics of photovoltaic and wind power, and the model has not undergone dynamic iterative optimization, resulting in large prediction errors. At the same time, dispatch control does not accurately allocate power based on equipment regulation capabilities and grid security constraints, and the compensation response to output fluctuations is lagging, which can easily cause problems such as curtailment of solar and wind power, overcharging and over-discharging of energy storage, or grid voltage and frequency fluctuations, seriously affecting the renewable energy absorption rate and system operation stability.
[0004] To address the aforementioned problems, a technical solution is provided. Summary of the Invention
[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a dynamic coordinated control method for photovoltaic and wind power complementary power generation systems and energy storage to solve the problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: A dynamic coordinated control method for photovoltaic-wind power complementary power generation systems and energy storage includes the following steps: Step S1: Collect new energy output data, environmental and load data, and energy storage status data; Step S2: Construct a hierarchical prediction model to predict the output trends of photovoltaic and wind power in the next 24 hours. Step S3: Combining the accurate output forecast value obtained from the hierarchical forecast with the remaining energy status of the energy storage, control commands are sent to the photovoltaic modules, wind turbines and energy storage system to adjust the output response of each device to match the load demand and grid operation requirements. Step S4: Monitor the deviation between the actual output of new energy and the predicted value in real time, and perform real-time energy storage fluctuation compensation when the deviation exceeds the preset threshold.
[0007] In a preferred embodiment, step S1 includes the following specific contents: Based on the power output fluctuation range of new energy units, and combined with the change cycle of environmental data, the time dimension of data collection is determined. The maximum coverage dimension of data acquisition is determined by using the real-time fluctuation range of load power as a constraint. Using the change threshold of energy storage SOC as the optimization condition, the time dimension of the collected data is calibrated to obtain a standardized multi-source data set.
[0008] In a preferred embodiment, based on a standardized multi-source dataset, the time matching deviation of each data dimension is calculated to generate a deviation correction matrix. Based on this correction matrix, the time dimension of the data is calibrated to obtain time-aligned multi-source data; With time alignment accuracy as the optimization goal, we conduct adaptation analysis on the data coverage dimensions, select the core data dimensions, and form a simplified and effective data set.
[0009] In a preferred embodiment, step S2 includes the following specific contents: Based on the streamlined and effective data set, a correlation model between new energy output and environmental and load data is constructed to generate a mapping relationship of multi-dimensional data. Based on this mapping relationship, the association paths between photovoltaic and wind power output and data dimensions are matched sequentially and marked as photovoltaic output association path and wind power output association path, respectively. By using the parameter sequence of the hierarchical prediction model, a predicted trajectory of the power output trend is generated based on the above-mentioned associated path, thus obtaining the photovoltaic and wind power output trend curves for the next 24 hours.
[0010] In a preferred embodiment, the feature dimensions of the data are extracted based on the fluctuation range of environmental data in the dataset; Based on this feature dimension, environmental data is input into the correlation model to obtain initial output trend prediction results; Based on the deviation between the prediction results and historical data, the parameters of the correlation model are iteratively corrected to obtain an optimized correlation model. The optimized correlation model is substituted into the hierarchical prediction model to update the output trend curve.
[0011] In a preferred embodiment, step S3 includes the following specific contents: First, obtain the accurate power output prediction value and the remaining energy storage power data obtained from the hierarchical prediction, and then divide the control parameters corresponding to different devices into zones. By calculating the difference between the predicted value and the actual load demand, and combining the output adjustment range of each device, the output adjustment rate of photovoltaic, wind power and energy storage is calculated respectively, and the target output range of each device is obtained. Based on the safety constraint model of power grid operation, the output adjustment margin of each device is extracted, and the gradient calculation method is used to obtain the output allocation weight of different devices. The target output range and output allocation weight are synchronized over time to generate control command parameters for each device.
[0012] In a preferred embodiment, a preset standard threshold for the power output stability of the power grid is established, and a benchmark curve for the power output stability is constructed using a curve fitting method. The actual output data of each device is compared with the reference curve to calculate the deviation between the current output and the reference curve, and the output deviation distribution is obtained for different devices. Align the output deviation distribution with the output regulation rate of each device, analyze the impact of output fluctuations on grid stability, and calculate the correlation coefficient between output deviation and regulation rate.
[0013] In a preferred embodiment, step S4 includes the following specific details: By comparing the actual output of new energy sources with the predicted values over a continuous period, and combining the monitoring time interval, the rate of change of output deviation is obtained, and the deviation data of all equipment is integrated into a deviation dataset. Based on the deviation dataset and the energy storage charging and discharging capacity model, the differential method is used to calculate the compensation power change rate of energy storage and determine the real-time compensation range of energy storage. By comparing the deviation data between the previous and subsequent cycles and combining it with the compensation response time, the trend of deviation change can be obtained.
[0014] The technical effects and advantages of the dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage in this invention are as follows: 1. By employing a multi-dimensional, hierarchical data acquisition and precise processing approach, the system ensures that data covers the core changing characteristics of new energy sources, the environment, load, and energy storage. It also resolves the issue of time misalignment in multi-source data, simplifies invalid data dimensions, and provides high-quality data support for subsequent prediction and control, significantly improving the overall accuracy of the control method. Simultaneously, the hierarchical prediction model, combined with the differentiated characteristics of photovoltaic and wind power, performs correlation path matching, greatly reducing output prediction errors caused by environmental fluctuations. 2. During the dispatching phase, precise power output allocation is carried out by combining equipment regulation capabilities with grid safety constraints, ensuring equipment operation safety and grid stability. During the fluctuation compensation phase, precise compensation of energy storage is achieved by monitoring the rate and trend of deviation changes in real time, effectively suppressing the impact of new energy output fluctuations on the grid, avoiding overcharging and over-discharging of energy storage, extending the service life of energy storage, and significantly improving the new energy consumption rate. Attached Figure Description
[0015] Figure 1 This is a flowchart illustrating the dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to the present invention. Detailed Implementation
[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example
[0017] Figure 1 The present invention provides a dynamic coordinated control method for a photovoltaic-wind power complementary power generation system and energy storage, which includes the following steps: Step S1: Collect new energy output data, environmental and load data, and energy storage status data; Step S2: Construct a hierarchical prediction model to predict the output trends of photovoltaic and wind power in the next 24 hours. Step S3: Combining the accurate output forecast value obtained from the hierarchical forecast with the remaining energy status of the energy storage, control commands are sent to the photovoltaic modules, wind turbines and energy storage system to adjust the output response of each device to match the load demand and grid operation requirements. Step S4: Monitor the deviation between the actual output of new energy and the predicted value in real time, and perform real-time energy storage fluctuation compensation when the deviation exceeds the preset threshold.
[0018] Step S1 includes the following specific contents: Based on the power output fluctuation range of new energy units, and combined with the change cycle of environmental data, the time dimension of data collection is determined. The maximum coverage dimension of data acquisition is determined by using the real-time fluctuation range of load power as a constraint. Using the change threshold of energy storage SOC as the optimization condition, the time dimension of the collected data is calibrated to obtain a standardized multi-source data set.
[0019] Based on a standardized multi-source dataset, the time matching deviation of each data dimension is calculated, and a deviation correction matrix is generated. Based on this correction matrix, the time dimension of the data is calibrated to obtain time-aligned multi-source data; With time alignment accuracy as the optimization goal, we conduct adaptation analysis on the data coverage dimensions, select the core data dimensions, and form a simplified and effective data set.
[0020] By using hierarchical constraints on output fluctuation range, load fluctuation range, and energy storage SOC threshold, it is ensured that data collection can fully cover the key changing characteristics of new energy, environment, load, and energy storage. Time matching deviation calculation and correction matrix calibration solve the time misalignment problem caused by the acquisition frequency and transmission delay of multi-source data, allowing data of each dimension to be accurately aligned on the time axis. The screening of core data dimensions simplifies the data volume, retains the most critical feature dimensions that affect the output of new energy, and improves data processing efficiency.
[0021] Step S2 includes the following specific contents: Based on the streamlined and effective data set, a correlation model between new energy output and environmental and load data is constructed to generate a mapping relationship of multi-dimensional data. Based on this mapping relationship, the association paths between photovoltaic and wind power output and data dimensions are matched sequentially and marked as photovoltaic output association path and wind power output association path, respectively. By using the parameter sequence of the hierarchical prediction model, a predicted trajectory of the power output trend is generated based on the above-mentioned associated path, thus obtaining the photovoltaic and wind power output trend curves for the next 24 hours.
[0022] Based on the fluctuation range of environmental data in the dataset, feature dimensions of the data are extracted; Based on this feature dimension, environmental data is input into the correlation model to obtain initial output trend prediction results; Based on the deviation between the prediction results and historical data, the parameters of the correlation model are iteratively corrected to obtain an optimized correlation model. The optimized correlation model is substituted into the hierarchical prediction model to update the output trend curve.
[0023] By constructing a correlation model between new energy output and multi-dimensional data, the mapping relationship between photovoltaic and wind power output and environmental and load data was clarified. The separate marking of the correlation paths of photovoltaic and wind power adapted to their different output characteristics, making the prediction more in line with the actual output patterns of each energy source. The extraction of environmental data feature dimensions and the iterative correction of the model can specifically adapt to the impact of environmental fluctuations on output. By optimizing the model parameters through deviation iteration, the prediction error caused by environmental changes was effectively reduced. The optimized model was substituted into hierarchical prediction, which ensured the global accuracy of the output trend in the next 24 hours.
[0024] Step S3 includes the following specific contents: First, obtain the accurate power output prediction value and the remaining energy storage power data obtained from the hierarchical prediction, and then divide the control parameters corresponding to different devices into zones. By calculating the difference between the predicted value and the actual load demand, and combining the output adjustment range of each device, the output adjustment rate of photovoltaic, wind power and energy storage is calculated respectively, and the target output range of each device is obtained. Based on the safety constraint model of power grid operation, the output adjustment margin of each device is extracted, and the gradient calculation method is used to obtain the output allocation weight of different devices. The target output range and output allocation weight are synchronized over time to generate control command parameters for each device.
[0025] A preset standard threshold for power output stability is established, and a benchmark curve for power output stability is constructed using curve fitting. The actual output data of each device is compared with the reference curve to calculate the deviation between the current output and the reference curve, and the output deviation distribution is obtained for different devices. Align the output deviation distribution with the output regulation rate of each device, analyze the impact of output fluctuations on grid stability, and calculate the correlation coefficient between output deviation and regulation rate.
[0026] By partitioning equipment control parameters, calculating output regulation rates, and determining target ranges, the output allocation of photovoltaic, wind power, and energy storage is made more aligned with their respective regulation capabilities, avoiding the problem of overload or insufficient regulation of individual equipment and ensuring the safety of each device's operation. Combining the power grid safety constraint model with gradient calculation of output allocation weights meets the requirements of power grid safety indicators such as voltage and frequency, maximizes the utilization of each device's regulation margin, and improves the overall output adaptation efficiency of the system. Through the construction and deviation analysis of output stability benchmark curves, the correlation between equipment output and power grid stability is accurately captured.
[0027] Step S4 includes the following specific contents: By comparing the actual output of new energy sources with the predicted values over a continuous period, and combining the monitoring time interval, the rate of change of output deviation is obtained, and the deviation data of all equipment is integrated into a deviation dataset. Based on the deviation dataset and the energy storage charging and discharging capacity model, the differential method is used to calculate the compensation power change rate of energy storage and determine the real-time compensation range of energy storage. By comparing the deviation data between the previous and subsequent cycles and combining it with the compensation response time, the trend of deviation change can be obtained.
[0028] By comparing the actual output with the predicted value over a continuous period, and by capturing the rate of change of deviation and integrating the deviation dataset through monitoring time intervals, the dynamic characteristics of new energy output fluctuations can be comprehensively and in real time grasped. Based on the deviation dataset and the energy storage charging and discharging capacity model, the rate of change of compensation power is calculated in a differential manner and the compensation range is determined to ensure that the energy storage compensation power matches its own charging and discharging capacity. This avoids overcharging and over-discharging of energy storage and damage to the equipment, while maximizing its compensation potential. By comparing the deviation data of previous and subsequent periods and analyzing the trend of change, combined with the compensation response time, the direction of deviation development can be predicted in advance and the compensation intensity can be adjusted to effectively suppress the continuous spread of output fluctuations.
[0029] In the data acquisition and preprocessing stage, the time dimension and coverage dimension of data acquisition are determined by using the fluctuation of new energy output, the range of load fluctuation, and the change of energy storage SOC as hierarchical constraints to avoid data redundancy while ensuring full coverage of core features. By calculating the time matching deviation of multi-source data and generating a correction matrix, the time alignment calibration of data is completed. Then, the core data dimensions are selected to form an effective dataset, which solves the problems of time misalignment and interference of invalid information in multi-source data.
[0030] In the power output prediction stage, a correlation model between new energy power output and multi-dimensional data is constructed based on the simplified effective dataset. For the differentiated characteristics of photovoltaic power's dependence on sunlight and wind power's dependence on wind speed, a dedicated power output correlation path is matched. Feature dimensions are extracted by combining the fluctuation range of environmental data. The environmental data is input into the correlation model to obtain the initial prediction results. Then, the model parameters are iteratively corrected by the deviation between the predicted value and historical data, and the power output trend curve of the hierarchical prediction model is updated to achieve accurate prediction of photovoltaic and wind power output in the next 24 hours.
[0031] In the coordinated scheduling phase, the control parameters of photovoltaic, wind power, and energy storage are first divided into zones based on the accurate output value predicted by the hierarchical forecast and the remaining energy storage capacity. By calculating the difference between the predicted output and the load demand and the equipment regulation rate, the target output range of each device is determined. Then, the equipment regulation margin is extracted based on the grid security constraint model, and the output allocation weight is obtained by gradient calculation. After time synchronization processing, control commands are generated to ensure that the output allocation conforms to the equipment regulation capacity and grid security requirements. At the same time, by constructing an output stability benchmark curve, the deviation between the actual output and the benchmark curve is compared, the correlation coefficient is calculated, and fine-tuning commands are generated to further optimize the scheduling accuracy.
[0032] In the fluctuation compensation stage, the deviation between the actual output of new energy and the predicted value is continuously monitored over a period of time. The rate of change of the deviation is obtained and the deviation dataset is integrated. Combined with the energy storage charging and discharging capacity model, the rate of change of compensation power and the real-time compensation range are calculated in a differential manner. By analyzing the changing trend of the deviation data in previous and subsequent periods, and combining the energy storage response time to predict the direction of deviation development, the energy storage compensation intensity is dynamically adjusted to achieve accurate offsetting of output fluctuations, ensure the stability of the total system output, and avoid the impact of fluctuations on the grid operation.
[0033] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0034] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A dynamic coordinated control method for a photovoltaic-wind power complementary power generation system and energy storage, characterized in that, Includes the following steps: Step S1: Collect new energy output data, environmental and load data, and energy storage status data; Step S2: Construct a hierarchical prediction model to predict the output trends of photovoltaic and wind power in the next 24 hours. Step S3: Combining the accurate output forecast value obtained from the hierarchical forecast with the remaining energy status of the energy storage, control commands are sent to the photovoltaic modules, wind turbines and energy storage system to adjust the output response of each device to match the load demand and grid operation requirements. Step S4: Monitor the deviation between the actual output of new energy and the predicted value in real time, and perform real-time energy storage fluctuation compensation when the deviation exceeds the preset threshold.
2. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 1, characterized in that: Step S1 includes the following specific contents: Based on the power output fluctuation range of new energy units, and combined with the change cycle of environmental data, the time dimension of data collection is determined. The maximum coverage dimension of data acquisition is determined by using the real-time fluctuation range of load power as a constraint. Using the change threshold of energy storage SOC as the optimization condition, the time dimension of the collected data is calibrated to obtain a standardized multi-source data set.
3. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 2, characterized in that: Based on a standardized multi-source dataset, the time matching deviation of each data dimension is calculated, and a deviation correction matrix is generated. Based on this correction matrix, the time dimension of the data is calibrated to obtain time-aligned multi-source data; With time alignment accuracy as the optimization goal, we conduct adaptation analysis on the data coverage dimensions, select the core data dimensions, and form a simplified and effective data set.
4. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 3, characterized in that: Step S2 includes the following specific contents: Based on the streamlined and effective data set, a correlation model between new energy output and environmental and load data is constructed to generate a mapping relationship of multi-dimensional data. Based on this mapping relationship, the association paths between photovoltaic and wind power output and data dimensions are matched sequentially and marked as photovoltaic output association path and wind power output association path, respectively. By using the parameter sequence of the hierarchical prediction model, a predicted trajectory of the power output trend is generated based on the above-mentioned associated path, thus obtaining the photovoltaic and wind power output trend curves for the next 24 hours.
5. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 4, characterized in that: Based on the fluctuation range of environmental data in the dataset, feature dimensions of the data are extracted; Based on this feature dimension, environmental data is input into the correlation model to obtain initial output trend prediction results; Based on the deviation between the prediction results and historical data, the parameters of the correlation model are iteratively corrected to obtain an optimized correlation model. The optimized correlation model is substituted into the hierarchical prediction model to update the output trend curve.
6. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 5, characterized in that: Step S3 includes the following specific contents: First, obtain the accurate power output prediction value and the remaining energy storage power data obtained from the hierarchical prediction, and then divide the control parameters corresponding to different devices into zones. By calculating the difference between the predicted value and the actual load demand, and combining the output adjustment range of each device, the output adjustment rate of photovoltaic, wind power and energy storage is calculated respectively, and the target output range of each device is obtained. Based on the safety constraint model of power grid operation, the output adjustment margin of each device is extracted, and the gradient calculation method is used to obtain the output allocation weight of different devices. The target output range and output allocation weight are synchronized over time to generate control command parameters for each device.
7. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 6, characterized in that: A preset standard threshold for power output stability is established, and a benchmark curve for power output stability is constructed using curve fitting. The actual output data of each device is compared with the reference curve to calculate the deviation between the current output and the reference curve, and the output deviation distribution is obtained for different devices. Align the output deviation distribution with the output regulation rate of each device, analyze the impact of output fluctuations on grid stability, and calculate the correlation coefficient between output deviation and regulation rate.
8. The dynamic coordinated control method for photovoltaic-wind power complementary power generation system and energy storage according to claim 7, characterized in that: Step S4 includes the following specific contents: By comparing the actual output of new energy sources with the predicted values over a continuous period, and combining the monitoring time interval, the rate of change of output deviation is obtained, and the deviation data of all equipment is integrated into a deviation dataset. Based on the deviation dataset and the energy storage charging and discharging capacity model, the differential method is used to calculate the compensation power change rate of energy storage and determine the real-time compensation range of energy storage. By comparing the deviation data between the previous and subsequent cycles and combining it with the compensation response time, the trend of deviation change can be obtained.