Optimized scheduling method of wind-solar-storage hybrid system

Abstract

The invention belongs to the technical field of wind-light storage hybrid systems, and particularly relates to an optimal scheduling method of a wind-light storage hybrid system, which comprises the following specific steps of: establishing a high-precision prediction model according to meteorological data, power generation data and power grid data so as to realize optimal scheduling of the wind-light storage hybrid system by utilizing historical data and a machine learning algorithm; the generated power of wind energy and solar energy is accurately predicted, and a basis is provided for a scheduling decision; according to a prediction result and a power grid load demand, making a power generation plan of the wind-light-storage hybrid system, and defining an output target of each power generation unit; dynamically adjusting the power distribution of each power generation unit by adopting an optimization algorithm according to the real-time wind and light power generation power and the energy storage state to ensure that the total output power of the system meets the requirements of a power grid; and formulating an energy storage charging and discharging strategy. Through reasonable scheduling, wind energy and solar energy resources are utilized to the maximum extent, the phenomenon of wind and light abandoning can be reduced, and the comprehensive utilization rate of energy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind-solar-storage hybrid systems, and specifically to an optimization scheduling method for wind-solar-storage hybrid systems. Background Art

[0002] As global demand for clean energy continues to grow, wind and solar energy, as important components of renewable energy, have been widely developed and utilized. However, wind and solar energy are intermittent and fluctuating, and their power generation is significantly affected by natural conditions, which poses a challenge to the stable operation of the power system. To effectively address this problem, wind-solar-storage hybrid systems have emerged. By combining wind power generation, photovoltaic power generation, and energy storage systems, they achieve energy complementarity and regulation, improving the reliability and stability of power supply. However, the system still has some scheduling issues, such as irrational energy distribution and low energy storage utilization efficiency. Therefore, an optimized scheduling method for a wind-solar-storage hybrid system is invented. Summary of the Invention

[0003] In view of the above problems and / or the problems existing in an existing optimization scheduling method for a wind-solar-storage hybrid system, the present invention is proposed.

[0004] Therefore, the purpose of the present invention is to provide an optimized scheduling method for a wind-solar-storage hybrid system, which can solve the problems of unreasonable energy distribution and low energy storage utilization efficiency during scheduling of the above-mentioned existing wind-solar-storage hybrid system.

[0005] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions:

[0006] A method for optimizing the scheduling of a wind-solar-storage hybrid system includes the following specific steps:

[0007] S1, Establishing a high-precision prediction model: Based on meteorological data, power generation data, and power grid data, a high-precision prediction model is established to accurately predict wind and solar power generation using historical data and machine learning algorithms, providing a basis for scheduling decisions;

[0008] S2. Develop a reasonable power generation plan: Based on the forecast results and grid load demand, develop a power generation plan for the wind-solar-storage hybrid system and clarify the output targets of each power generation unit;

[0009] S3, power allocation optimization: Based on the real-time wind and solar power generation and energy storage status, an optimization algorithm is used to dynamically adjust the power allocation of each power generation unit to ensure that the total system output power meets the grid demand;

[0010] S4, energy storage charge and discharge control: Develop energy storage charge and discharge strategies to charge when wind and photovoltaic power are in excess and discharge when power is insufficient, thus achieving peak load shifting and valley filling.

[0011] S5, Introducing Intelligent Algorithms: Introducing deep reinforcement learning algorithms to enable the system to autonomously learn and make optimal scheduling decisions based on real-time operating status and historical experience;

[0012] S6, Multi-scenario Simulation and Verification: Based on historical data and forecast results, we first construct a variety of complex operating scenarios, including different seasons, weather conditions, peak and valley periods, and possible emergencies. These scenarios are then simulated and run. The simulation results are then compared with the expected goals to evaluate the effectiveness and adaptability of the scheduling strategy. Problems encountered during the simulation can be analyzed, and algorithms and strategies can be adjusted accordingly.

[0013] S7, establish a decision support system: integrate various data information, provide dispatchers with a visual decision interface, and assist them in making scientific decisions;

[0014] S8, real-time model verification and update: First, operational data from wind farms, photovoltaic power plants, and energy storage systems is collected in real time. This data is then compared and analyzed with the data in the prediction model and scheduling strategy. The accuracy of the prediction model is then evaluated using statistical methods and machine learning techniques. If a significant deviation between the model prediction results and the actual data is found, the prediction model update program is immediately initiated to optimize the prediction model and scheduling algorithm based on the latest data to ensure that they always match the actual system operating status.

[0015] S9, Equipment Health Status Assessment and Preventive Maintenance: First, use sensors to collect the operating parameters of the equipment in real time. Then, use big data analysis technology and equipment failure prediction models to assess the health status of the equipment and predict possible equipment failures. Then, based on the equipment health status assessment results, formulate a preventive maintenance plan, reasonably arrange the equipment inspection and maintenance time, and replace parts that may fail in advance. In addition, the equipment health status information can be integrated into the decision support system to provide a reference for scheduling decisions. When the equipment health status is poor, the scheduling strategy is adjusted to avoid abnormal system operation due to equipment failure.

[0016] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S1 are as follows:

[0017] S11, data collection: collect historical meteorological data, power generation data and power grid data;

[0018] S12, Data Preprocessing: Data cleaning, data standardization and data feature engineering are performed on the collected data. The data feature engineering is to extract and combine features of the original data based on business knowledge and data analysis results to create new features for mining the potential value of the data and improving the model prediction ability;

[0019] S13, model building: building a deep learning model;

[0020] S14, model training: training the constructed deep learning model.

[0021] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S14 are as follows:

[0022] S141: First, the collected data is divided into a training set, a validation set, and a test set in a ratio of 70%, 15%, and 15%. The training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the generalization ability of the model.

[0023] S142: Set the corresponding parameters according to the model type and data characteristics, and optimize the parameters through grid search and random search methods to find the optimal parameter combination;

[0024] S143: Use the training set data to train the selected model, and continuously adjust the model parameters so that the model can learn the patterns and features in the data. During the training process, early stopping can be used to avoid overfitting of the model.

[0025] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S2 are as follows:

[0026] S21, load forecasting: first collect load data, then perform load forecasting based on the load data;

[0027] S22, Wind, Solar, and Storage Resource Assessment: First, forecast the future power generation of wind farms and photovoltaic power stations and analyze the uncertainty of the forecast results. Then, monitor the energy storage system's state of charge, charge and discharge power limits, and charge and discharge efficiency parameters in real time to determine the energy storage system's available capacity and regulation capabilities at different times.

[0028] S23, target setting and constraint determination: First, determine the overall requirements of system operation and set the target of power generation plan, then clarify the constraint conditions;

[0029] S24, power generation plan optimization model construction and solution: first, based on the load forecast, resource assessment results, and objectives and constraints, a mathematical optimization model is constructed, and then a corresponding solution algorithm is selected for the constructed mathematical optimization model;

[0030] S25, Plan Formulation and Adjustment: First, the optimization model is solved using the selected solution algorithm to obtain the power generation plan for each power generation unit in each future time period. The generated power generation plan is then checked for rationality to analyze whether there are equipment overloads or insufficient energy storage capacity. If any problems are found, the model parameters or constraints are adjusted and the model is re-solved. After that, a rolling optimization method is used to regularly update the load forecast and wind and solar power generation forecast data, re-solve the optimization model, and dynamically adjust the power generation plan to ensure that the plan always adapts to actual operating conditions.

[0031] S26, Plan Execution and Feedback: First, the finalized power generation plan is issued to the control systems of each power generation unit and energy storage system to ensure smooth execution of the plan. Then, during the execution of the power generation plan, the actual operation of each power generation unit and energy storage system is monitored in real time, and operation data is collected. The actual operation data is compared with the planned data, and the cause of the deviation is analyzed. If the deviation is large, timely adjustment measures are taken, and a reference is provided for the optimization of subsequent power generation plans.

[0032] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S3 are as follows:

[0033] S31, real-time data collection: real-time collection of power generation unit data, energy storage system data and grid load data;

[0034] S32, Data Analysis and Forecasting: First, combine historical data and real-time monitoring information to forecast the future power generation of wind farms and photovoltaic power stations. Then, conduct a short-term forecast of grid load and analyze load trends. Finally, assess the uncertainty of wind and solar power generation forecasts and load forecasts, quantifying the scope and degree of uncertainty. By considering these uncertainties, the power allocation plan can be made more robust and adaptable to various changes in actual operation.

[0035] S33, determination of optimization objectives and constraints: first set the optimization objectives, then clarify the constraints;

[0036] S34, power allocation model construction: first determine the decision variables, then build the objective function, and then establish the constraint equations;

[0037] S35, model solution: First, select the corresponding solution algorithm based on the characteristics of the power distribution model. Then, use the selected solution algorithm to solve the power distribution model to obtain the optimal power distribution plan for each power generation unit and energy storage system in different time periods.

[0038] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S33 are as follows:

[0039] S331, setting optimization target: determining the optimization target of power distribution according to the system operation requirements and actual needs;

[0040] S332, clarify the constraints:

[0041] S3321, Power Balance Constraint: Ensures that the system's total generated power matches the grid load demand, while considering the impact of the energy storage system's charge and discharge power on power balance;

[0042] S3322, Equipment Operation Constraints: Comply with the power limits and charge / discharge cycle limits of wind turbines, photovoltaic panels, and energy storage systems to prevent equipment overload and extend equipment life.

[0043] S3323, Grid Access Constraints: Meet the grid's requirements for voltage, frequency, and power factor to ensure a safe and stable connection between the system and the grid.

[0044] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S34 are as follows:

[0045] S341, determining decision variables: using the generated power of the wind farm and photovoltaic power station in each time period, as well as the charge and discharge power of the energy storage system, as decision variables, and achieving optimal power allocation by reasonably adjusting the values ​​of these variables;

[0046] S342, constructing an objective function: constructing a corresponding objective function according to the set optimization goal;

[0047] S343, establishing constraint equations: expressing the determined constraint conditions in the form of equations or inequalities as constraint equations of the model, to ensure that the power allocation scheme is optimized under the premise of satisfying various conditions.

[0048] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S4 are as follows:

[0049] S41, Data Collection and Integration: Collect and integrate energy generation data, grid load data, and energy storage system data;

[0050] S42, system status assessment: first perform power balance analysis, then perform energy storage status assessment;

[0051] S43, charge and discharge strategy formulation: first implement a strategy based on power balance, then implement a strategy based on electricity price;

[0052] S44, charge and discharge control model construction: First, the charge and discharge power of the energy storage system is used as the control variable. By adjusting the size and time of the charge and discharge power, the energy storage system can be effectively controlled. Then, according to the established charge and discharge strategy, the corresponding objective function is constructed. Finally, the physical characteristics and operating limitations of the energy storage system are considered and the constraints are set.

[0053] S45, model solution and instruction generation: First, select the corresponding solution algorithm based on the characteristics of the charge and discharge control model, then use the selected solution algorithm to solve the charge and discharge control model to obtain the optimal charge and discharge power of the energy storage system in different time periods, and generate the corresponding charge and discharge control instructions based on the solution results.

[0054] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S42 are as follows:

[0055] S421, Power Balance Analysis: Compares generated power with grid load to determine whether the system is in a power surplus, power deficit, or power balance state. If generated power exceeds the load demand, the system is in a power surplus state and can charge the energy storage system. If generated power is less than the load demand, the system is in a power deficit state and can discharge the energy storage system to supplement power.

[0056] S422, Energy Storage Status Assessment: Based on the collected energy storage system data, comprehensively assess its operating status.

[0057] As a preferred solution of the optimized scheduling method of a wind-solar-storage hybrid system described in the present invention, the specific steps of S43 are as follows:

[0058] S431, based on power balancing strategy: When there is excess power, the energy storage system is prioritized for charging and storing the excess energy; when there is insufficient power, the energy storage system is controlled to discharge and replenish power to the grid. A power threshold can be set, and when the difference between the generated power and the load power exceeds the threshold, the corresponding charging and discharging operations are initiated;

[0059] S432, electricity price-based strategy: Arranges energy storage system charging during low-price periods; discharges during peak-price periods, shaving peaks and filling valleys, reducing system operating costs. Analyzes historical and real-time electricity price data to develop reasonable charging and discharging time and power plans.

[0060] S433, a strategy that comprehensively considers multiple factors: comprehensively considers factors such as power balance, energy storage status, and electricity price, and uses optimization algorithms to formulate the optimal charging and discharging strategy.

[0061] Compared with existing technologies:

[0062] 1. The present invention maximizes the use of wind and solar energy resources through reasonable scheduling, which can reduce the phenomenon of wind and solar power abandonment and improve the comprehensive utilization rate of energy.

[0063] 2. The present invention utilizes the regulating function of the energy storage system to smooth out the power fluctuations of wind power and photovoltaic power, ensure the stability of power output, and reduce the impact on the power grid.

[0064] 3. By optimizing the energy distribution strategy, the present invention can reduce the number of charge and discharge cycles of the energy storage system, extend the life of the energy storage equipment, and reduce the overall operating cost of the system. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below.

[0066] The present invention provides an optimization scheduling method for a wind-solar-storage hybrid system, comprising the following specific steps:

[0067] S1, Establishing a high-precision prediction model: Based on meteorological data, power generation data, and power grid data, a high-precision prediction model is established to accurately predict wind and solar power generation using historical data and machine learning algorithms, providing a basis for scheduling decisions;

[0068] The specific steps of S1 are as follows:

[0069] S11, data collection: collect historical meteorological data, power generation data and power grid data;

[0070] S12, Data Preprocessing: Data cleaning, data standardization and data feature engineering are performed on the collected data. The data feature engineering is to extract and combine features of the original data based on business knowledge and data analysis results to create new features for mining the potential value of the data and improving the model prediction ability;

[0071] S13, model building: building a deep learning model;

[0072] S14, model training: training the constructed deep learning model;

[0073] The specific steps of S14 are as follows:

[0074] S141: First, the collected data is divided into a training set, a validation set, and a test set in a ratio of 70%, 15%, and 15%. The training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the generalization ability of the model.

[0075] S142: Set the corresponding parameters according to the model type and data characteristics, and optimize the parameters through grid search and random search methods to find the optimal parameter combination;

[0076] S143: Use the training set data to train the selected model, and continuously adjust the model parameters so that the model can learn the patterns and characteristics in the data. During the training process, early stopping can be used to avoid overfitting of the model.

[0077] S2. Develop a reasonable power generation plan: Based on the forecast results and grid load demand, develop a power generation plan for the wind-solar-storage hybrid system and clarify the output targets of each power generation unit;

[0078] The specific steps of S2 are as follows:

[0079] S21, load forecasting: first collect load data, then perform load forecasting based on the load data;

[0080] S22, Wind, Solar, and Storage Resource Assessment: First, forecast the future power generation of wind farms and photovoltaic power stations and analyze the uncertainty of the forecast results. Then, monitor the energy storage system's state of charge, charge and discharge power limits, and charge and discharge efficiency parameters in real time to determine the energy storage system's available capacity and regulation capabilities at different times.

[0081] S23, target setting and constraint determination: First, determine the overall requirements of system operation and set the target of power generation plan, then clarify the constraint conditions;

[0082] S24, power generation plan optimization model construction and solution: first, based on the load forecast, resource assessment results, and objectives and constraints, a mathematical optimization model is constructed, and then a corresponding solution algorithm is selected for the constructed mathematical optimization model;

[0083] S25, Plan Formulation and Adjustment: First, the optimization model is solved using the selected solution algorithm to obtain the power generation plan for each power generation unit in each future time period. The generated power generation plan is then checked for rationality to analyze whether there are equipment overloads or insufficient energy storage capacity. If any problems are found, the model parameters or constraints are adjusted and the model is re-solved. After that, a rolling optimization method is used to regularly update the load forecast and wind and solar power generation forecast data, re-solve the optimization model, and dynamically adjust the power generation plan to ensure that the plan always adapts to actual operating conditions.

[0084] S26, Plan Execution and Feedback: First, the finalized power generation plan is issued to the control systems of each power generation unit and energy storage system to ensure smooth execution. Then, during the execution of the power generation plan, the actual operation of each power generation unit and energy storage system is monitored in real time, and operating data is collected. The actual operating data is compared with the planned data, and the cause of any deviation is analyzed. If the deviation is large, timely adjustment measures are taken, and this provides reference for the optimization of subsequent power generation plans.

[0085] S3, power allocation optimization: Based on the real-time wind and solar power generation and energy storage status, an optimization algorithm is used to dynamically adjust the power allocation of each power generation unit to ensure that the total system output power meets the grid demand;

[0086] The specific steps of S3 are as follows:

[0087] S31, real-time data collection: real-time collection of power generation unit data, energy storage system data and grid load data;

[0088] S32, Data Analysis and Forecasting: First, combine historical data and real-time monitoring information to forecast the future power generation of wind farms and photovoltaic power stations. Then, conduct a short-term forecast of grid load and analyze load trends. Finally, assess the uncertainty of wind and solar power generation forecasts and load forecasts, quantifying the scope and degree of uncertainty. By considering these uncertainties, the power allocation plan can be made more robust and adaptable to various changes in actual operation.

[0089] S33, determination of optimization objectives and constraints: first set the optimization objectives, then clarify the constraints;

[0090] The specific steps of S33 are as follows:

[0091] S331, setting optimization target: determining the optimization target of power distribution according to the system operation requirements and actual needs;

[0092] S332, clarify the constraints:

[0093] S3321, Power Balance Constraint: Ensures that the system's total generated power matches the grid load demand, while considering the impact of the energy storage system's charge and discharge power on power balance;

[0094] S3322, Equipment Operation Constraints: Comply with the power limits and charge / discharge cycle limits of wind turbines, photovoltaic panels, and energy storage systems to prevent equipment overload and extend equipment life.

[0095] S3323, Grid Access Constraints: Meet the grid's requirements for voltage, frequency, and power factor to ensure a secure and stable connection between the system and the grid;

[0096] S34, power allocation model construction: first determine the decision variables, then build the objective function, and then establish the constraint equations;

[0097] The specific steps of S34 are as follows:

[0098] S341, determining decision variables: using the generated power of the wind farm and photovoltaic power station in each time period, as well as the charge and discharge power of the energy storage system, as decision variables, and achieving optimal power allocation by reasonably adjusting the values ​​of these variables;

[0099] S342, constructing an objective function: constructing a corresponding objective function according to the set optimization goal;

[0100] S343, establishing constraint equations: Expressing the determined constraint conditions in the form of equations or inequalities as constraint equations of the model to ensure that the power allocation scheme is optimized while satisfying various conditions;

[0101] S35, model solution: First, select a corresponding solution algorithm based on the characteristics of the power distribution model. Then, use the selected solution algorithm to solve the power distribution model to obtain the optimal power distribution plan for each power generation unit and energy storage system at different time periods;

[0102] S4, energy storage charge and discharge control: Develop energy storage charge and discharge strategies to charge when wind and photovoltaic power are in excess and discharge when power is insufficient, thus achieving peak load shifting and valley filling.

[0103] The specific steps of S4 are as follows:

[0104] S41, Data Collection and Integration: Collect and integrate energy generation data, grid load data, and energy storage system data;

[0105] S42, system status assessment: first perform power balance analysis, then perform energy storage status assessment;

[0106] The specific steps of S42 are as follows:

[0107] S421, Power Balance Analysis: Compares generated power with grid load to determine whether the system is in a power surplus, power deficit, or power balance state. If generated power exceeds the load demand, the system is in a power surplus state and can charge the energy storage system. If generated power is less than the load demand, the system is in a power deficit state and can discharge the energy storage system to supplement power.

[0108] S422, Energy Storage Status Assessment: Based on the collected energy storage system data, comprehensively assess its operating status;

[0109] S43, charge and discharge strategy formulation: first implement a strategy based on power balance, then implement a strategy based on electricity price;

[0110] The specific steps of S43 are as follows:

[0111] S431, based on power balancing strategy: When there is excess power, the energy storage system is prioritized for charging and storing the excess energy; when there is insufficient power, the energy storage system is controlled to discharge and replenish power to the grid. A power threshold can be set, and when the difference between the generated power and the load power exceeds the threshold, the corresponding charging and discharging operations are initiated;

[0112] S432, electricity price-based strategy: Arranges energy storage system charging during low-price periods; discharges during peak-price periods, shaving peaks and filling valleys, reducing system operating costs. Analyzes historical and real-time electricity price data to develop reasonable charging and discharging time and power plans.

[0113] S433, a strategy that comprehensively considers multiple factors: comprehensively considers factors such as power balance, energy storage status, and electricity price, and uses optimization algorithms to formulate the optimal charging and discharging strategy;

[0114] S44, charge and discharge control model construction: First, the charge and discharge power of the energy storage system is used as the control variable. By adjusting the size and time of the charge and discharge power, the energy storage system can be effectively controlled. Then, according to the established charge and discharge strategy, the corresponding objective function is constructed. Finally, the physical characteristics and operating limitations of the energy storage system are considered and the constraints are set.

[0115] S45, model solution and instruction generation: First, a corresponding solution algorithm is selected based on the characteristics of the charge and discharge control model. Then, the selected solution algorithm is used to solve the charge and discharge control model to obtain the optimal charge and discharge power of the energy storage system at different time periods. Based on the solution results, corresponding charge and discharge control instructions are generated;

[0116] S5, Introducing Intelligent Algorithms: Introducing deep reinforcement learning algorithms to enable the system to autonomously learn and make optimal scheduling decisions based on real-time operating status and historical experience;

[0117] S6, Multi-scenario Simulation and Verification: Based on historical data and forecast results, we first construct a variety of complex operating scenarios, including different seasons, weather conditions, peak and valley periods, and possible emergencies. These scenarios are then simulated and run. The simulation results are then compared with the expected goals to evaluate the effectiveness and adaptability of the scheduling strategy. Problems encountered during the simulation can be analyzed, and algorithms and strategies can be adjusted accordingly.

[0118] By setting up multi-scenario simulation and verification, we can comprehensively test the performance of the scheduling strategy under different working conditions, identify and resolve potential problems in advance, enhance the robustness and flexibility of the scheduling strategy, and ensure that the system can operate stably under various complex situations;

[0119] S7, establish a decision support system: integrate various data information, provide dispatchers with a visual decision interface, and assist them in making scientific decisions;

[0120] S8, real-time model verification and update: First, operational data from wind farms, photovoltaic power plants, and energy storage systems is collected in real time. This data is then compared and analyzed with the data in the prediction model and scheduling strategy. The accuracy of the prediction model is then evaluated using statistical methods and machine learning techniques. If a significant deviation between the model prediction results and the actual data is found, the prediction model update program is immediately initiated to optimize the prediction model and scheduling algorithm based on the latest data to ensure that they always match the actual system operating status.

[0121] By setting up real-time model verification and updating, the timeliness and accuracy of the prediction model and scheduling algorithm can be guaranteed, enabling the system to make optimal decisions based on real-time changes, improving the system's adaptability and further improving energy efficiency and system stability.

[0122] S9, Equipment Health Assessment and Preventive Maintenance: First, use sensors to collect equipment operating parameters in real time. Then, apply big data analysis technology and equipment failure prediction models to assess the equipment's health status and predict potential equipment failures. Based on the equipment health assessment results, a preventive maintenance plan is developed to rationally schedule equipment inspection and maintenance, and replace potentially faulty parts in advance. Equipment health information can also be integrated into the decision support system to provide a reference for scheduling decisions. When equipment health is poor, the scheduling strategy is adjusted to avoid system operation abnormalities caused by equipment failures.

[0123] By setting up equipment health status assessment and preventive maintenance, the impact of sudden equipment failures on system operation can be reduced, the equipment service life can be extended, the equipment maintenance cost can be reduced, and the continuous and stable operation of the system can be guaranteed.

[0124] Although the present invention has been described above with reference to embodiments, various modifications may be made thereto and equivalent components may be substituted without departing from the scope of the present invention. In particular, as long as there are no structural conflicts, the various features of the embodiments disclosed herein may be combined with each other in any manner, and the omission of an exhaustive description of such combinations in this specification is solely for the sake of space and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. An optimization scheduling method for a wind-solar-storage hybrid system, characterized in that: The specific steps are as follows: S1, Establishing a high-precision prediction model: Based on meteorological data, power generation data, and power grid data, a high-precision prediction model is established to accurately predict wind and solar power generation using historical data and machine learning algorithms, providing a basis for scheduling decisions; S2. Develop a reasonable power generation plan: Based on the forecast results and grid load demand, develop a power generation plan for the wind-solar-storage hybrid system and clarify the output targets of each power generation unit; S3, power allocation optimization: Based on the real-time wind and solar power generation and energy storage status, an optimization algorithm is used to dynamically adjust the power allocation of each power generation unit to ensure that the total system output power meets the grid demand; S4, energy storage charge and discharge control: Develop energy storage charge and discharge strategies to charge when wind and photovoltaic power are in excess and discharge when power is insufficient, thus achieving peak load shifting and valley filling. S5, Introducing Intelligent Algorithms: Introducing deep reinforcement learning algorithms to enable the system to autonomously learn and make optimal scheduling decisions based on real-time operating status and historical experience; S6, Multi-scenario Simulation and Verification: Based on historical data and forecast results, we first construct a variety of complex operating scenarios, including different seasons, weather conditions, peak and valley periods, and possible emergencies. These scenarios are then simulated and run. The simulation results are then compared with the expected goals to evaluate the effectiveness and adaptability of the scheduling strategy. Problems encountered during the simulation can be analyzed, and algorithms and strategies can be adjusted accordingly. S7, establish a decision support system: integrate various data information, provide dispatchers with a visual decision interface, and assist them in making scientific decisions; S8, real-time model verification and update: First, operational data from wind farms, photovoltaic power plants, and energy storage systems is collected in real time. This data is then compared and analyzed with the data in the prediction model and scheduling strategy. The accuracy of the prediction model is then evaluated using statistical methods and machine learning techniques. If a significant deviation between the model prediction results and the actual data is found, the prediction model update program is immediately initiated to optimize the prediction model and scheduling algorithm based on the latest data to ensure that they always match the actual system operating status. S9, Equipment Health Status Assessment and Preventive Maintenance: First, use sensors to collect the operating parameters of the equipment in real time. Then, use big data analysis technology and equipment failure prediction models to assess the health status of the equipment and predict possible equipment failures. Then, based on the equipment health status assessment results, formulate a preventive maintenance plan, reasonably arrange the equipment inspection and maintenance time, and replace parts that may fail in advance. In addition, the equipment health status information can be integrated into the decision support system to provide a reference for scheduling decisions. When the equipment health status is poor, the scheduling strategy is adjusted to avoid abnormal system operation due to equipment failure.

2. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 1 is characterized in that: The specific steps of S1 are as follows: S11, data collection: collect historical meteorological data, power generation data and power grid data; S12, Data Preprocessing: Data cleaning, data standardization and data feature engineering are performed on the collected data. The data feature engineering is to extract and combine features of the original data based on business knowledge and data analysis results to create new features for mining the potential value of the data and improving the model prediction ability; S13, model building: building a deep learning model; S14, model training: training the constructed deep learning model.

3. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 2 is characterized in that: The specific steps of S14 are as follows: S141: First, the collected data is divided into a training set, a validation set, and a test set in a ratio of 70%, 15%, and 15%. The training set is used to train the model, the validation set is used to adjust the model parameters, and the test set is used to evaluate the generalization ability of the model. S142: Set the corresponding parameters according to the model type and data characteristics, and optimize the parameters through grid search and random search methods to find the optimal parameter combination; S143: Use the training set data to train the selected model, and continuously adjust the model parameters so that the model can learn the patterns and features in the data. During the training process, early stopping can be used to avoid overfitting of the model.

4. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 1, characterized in that: The specific steps of S2 are as follows: S21, load forecasting: first collect load data, then perform load forecasting based on the load data; S22, Wind, Solar, and Storage Resource Assessment: First, forecast the future power generation of wind farms and photovoltaic power stations and analyze the uncertainty of the forecast results. Then, monitor the energy storage system's state of charge, charge and discharge power limits, and charge and discharge efficiency parameters in real time to determine the energy storage system's available capacity and regulation capabilities at different times. S23, target setting and constraint determination: First, determine the overall requirements of system operation and set the target of power generation plan, then clarify the constraint conditions; S24, power generation plan optimization model construction and solution: first, based on the load forecast, resource assessment results, and objectives and constraints, a mathematical optimization model is constructed, and then a corresponding solution algorithm is selected for the constructed mathematical optimization model; S25, Plan Formulation and Adjustment: First, the optimization model is solved using the selected solution algorithm to obtain the power generation plan for each power generation unit in each future time period. The generated power generation plan is then checked for rationality to analyze whether there are equipment overloads or insufficient energy storage capacity. If any problems are found, the model parameters or constraints are adjusted and the model is re-solved. After that, a rolling optimization method is used to regularly update the load forecast and wind and solar power generation forecast data, re-solve the optimization model, and dynamically adjust the power generation plan to ensure that the plan always adapts to actual operating conditions. S26, Plan Execution and Feedback: First, the finalized power generation plan is issued to the control systems of each power generation unit and energy storage system to ensure smooth execution of the plan. Then, during the execution of the power generation plan, the actual operation of each power generation unit and energy storage system is monitored in real time, and operation data is collected. The actual operation data is compared with the planned data, and the cause of the deviation is analyzed. If the deviation is large, timely adjustment measures are taken, and a reference is provided for the optimization of subsequent power generation plans.

5. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 1, characterized in that: The specific steps of S3 are as follows: S31, real-time data collection: real-time collection of power generation unit data, energy storage system data and grid load data; S32, Data Analysis and Forecasting: First, combine historical data and real-time monitoring information to forecast the future power generation of wind farms and photovoltaic power stations. Then, conduct a short-term forecast of grid load and analyze load trends. Finally, assess the uncertainty of wind and solar power generation forecasts and load forecasts, quantifying the scope and degree of uncertainty. By considering these uncertainties, the power allocation plan can be made more robust and adaptable to various changes in actual operation. S33, determination of optimization objectives and constraints: first set the optimization objectives, then clarify the constraints; S34, power allocation model construction: first determine the decision variables, then build the objective function, and then establish the constraint equations; S35, model solution: First, select the corresponding solution algorithm based on the characteristics of the power distribution model. Then, use the selected solution algorithm to solve the power distribution model to obtain the optimal power distribution plan for each power generation unit and energy storage system in different time periods.

6. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 5 is characterized in that: The specific steps of S33 are as follows: S331, setting optimization target: determining the optimization target of power distribution according to the system operation requirements and actual needs; S332, clarify the constraints: S3321, Power Balance Constraint: Ensures that the system's total generated power matches the grid load demand, while considering the impact of the energy storage system's charge and discharge power on power balance; S3322, Equipment Operation Constraints: Comply with the power limits and charge / discharge cycle limits of wind turbines, photovoltaic panels, and energy storage systems to prevent equipment overload and extend equipment life. S3323, Grid Access Constraints: Meet the grid's requirements for voltage, frequency, and power factor to ensure a safe and stable connection between the system and the grid.

7. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 5, characterized in that: The specific steps of S34 are as follows: S341, determining decision variables: using the generated power of the wind farm and photovoltaic power station in each time period, as well as the charge and discharge power of the energy storage system, as decision variables, and achieving optimal power allocation by reasonably adjusting the values ​​of these variables; S342, constructing an objective function: constructing a corresponding objective function according to the set optimization goal; S343, establishing constraint equations: expressing the determined constraint conditions in the form of equations or inequalities as constraint equations of the model, to ensure that the power allocation scheme is optimized under the premise of satisfying various conditions.

8. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 1, characterized in that: The specific steps of S4 are as follows: S41, Data Collection and Integration: Collect and integrate energy generation data, grid load data, and energy storage system data; S42, system status assessment: first perform power balance analysis, then perform energy storage status assessment; S43, charge and discharge strategy formulation: first implement a strategy based on power balance, then implement a strategy based on electricity price; S44, charge and discharge control model construction: First, the charge and discharge power of the energy storage system is used as the control variable. By adjusting the size and time of the charge and discharge power, the energy storage system can be effectively controlled. Then, according to the established charge and discharge strategy, the corresponding objective function is constructed. Finally, the physical characteristics and operating limitations of the energy storage system are considered and the constraints are set. S45, model solution and instruction generation: First, select the corresponding solution algorithm based on the characteristics of the charge and discharge control model, then use the selected solution algorithm to solve the charge and discharge control model to obtain the optimal charge and discharge power of the energy storage system in different time periods, and generate the corresponding charge and discharge control instructions based on the solution results.

9. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 8, characterized in that: The specific steps of S42 are as follows: S421, Power Balance Analysis: Compares generated power with grid load to determine whether the system is in a power surplus, power deficit, or power balance state. If generated power exceeds the load demand, the system is in a power surplus state and can charge the energy storage system. If generated power is less than the load demand, the system is in a power deficit state and can discharge the energy storage system to supplement power. S422, Energy Storage Status Assessment: Based on the collected energy storage system data, comprehensively assess its operating status.

10. The optimization scheduling method of a wind-solar-storage hybrid system according to claim 8, characterized in that: The specific steps of S43 are as follows: S431, based on power balancing strategy: When there is excess power, the energy storage system is prioritized for charging and storing the excess energy; when there is insufficient power, the energy storage system is controlled to discharge and replenish power to the grid. A power threshold can be set, and when the difference between the generated power and the load power exceeds the threshold, the corresponding charging and discharging operations are initiated; S432, electricity price-based strategy: Arranges energy storage system charging during low-price periods; discharges during peak-price periods, shaving peaks and filling valleys, reducing system operating costs. Analyzes historical and real-time electricity price data to develop reasonable charging and discharging time and power plans. S433, a strategy that comprehensively considers multiple factors: comprehensively considers factors such as power balance, energy storage status, and electricity price, and uses optimization algorithms to formulate the optimal charging and discharging strategy.