Energy storage regulation and control method and system for virtual power plant

Abstract

The invention discloses an energy storage regulation and control method and system for a virtual power plant, and the method comprises the following steps: S1, data collection: a metering sensor and metering equipment are installed at DERs to collect the data of the generating capacity, the energy storage state and the load state, the DERs are various distributed energy equipment or systems forming the virtual power plant, and the DERs are distributed energy equipment or systems forming the virtual power plant; a wired or wireless communication network is configured to connect the metering sensor and the metering device to ensure that data can be transmitted from the DERs to the central control system. The method has the advantages of high efficiency and stability, and solves the problems that in the energy storage regulation process of an existing virtual power plant, efficient and reliable energy storage regulation is inconvenient to carry out, unbalance of power regulation is easily caused, maximization of economic benefits and environmental protection benefits are affected, and stability of energy storage regulation is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to an energy storage control method and system for virtual power plants. Background Art

[0002] A virtual power plant is an intelligent management system that integrates and coordinates multiple distributed energy resources. It is not a physical power plant, but rather connects decentralized small-scale power generation facilities, energy storage systems, adjustable loads, and other energy facilities through advanced information technology and communication technology, and participates in the electricity market as a whole.

[0003] A method, device and virtual power plant for energy storage control with Chinese patent application number 201911293640.5, wherein the method is applied to a controller of a virtual power plant, and the virtual power plant also includes: a primary energy storage system and a secondary energy storage system, wherein the primary energy storage system is connected to the distribution system of the virtual power plant, and the secondary energy storage system is deployed in a park microgrid within the virtual power plant, and when the operation mode of the virtual power plant is determined, the power output from the park microgrid connection point to the distribution system is controlled within a controllable range under the operation mode by controlling the charging or discharging of the secondary energy storage system, and the output power of the distribution system is kept within a stable range by controlling the primary energy storage system to operate in a peak shaving and valley filling mode, thereby ensuring the stability of the virtual power plant when participating in power grid operation or power trading.

[0004] The existing virtual power plants are not conducive to efficient and reliable energy storage regulation during the energy storage regulation process, which can easily lead to an imbalance in power regulation, affect the maximization of economic benefits and environmental benefits, and reduce the stability of energy storage regulation. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for energy storage control for a virtual power plant, which has the advantages of high efficiency and stability, and solves the problem that the existing virtual power plant is not convenient for efficient and reliable energy storage control during the process of energy storage control, which easily leads to an imbalance in power control, affects the maximization of economic benefits and environmental benefits, and reduces the stability of energy storage control.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for regulating energy storage in a virtual power plant, comprising the following steps:

[0007] S1. Data collection: Metering sensors and metering equipment are installed at DERs to collect data on power generation, energy storage status, and load status. DERs are various distributed energy devices or systems that make up the virtual power plant. Wired or wireless communication networks are configured to connect metering sensors and metering equipment to ensure that data can be transmitted from DERs to the central control system.

[0008] S2. Edge computing deployment: Edge computing devices are installed at data collection points. The central control system transmits the received data to the edge computing devices. The edge computing devices perform preliminary data preprocessing. During preprocessing, data cleaning tools are used to remove outliers and fill in missing values ​​to ensure data quality. The data is standardized so that data from different sources can be compared under the same dimension.

[0009] S3 data integration, first build a data warehouse to store historical data and real-time data, and then use the data fusion platform to integrate data from different sources into a unified data set;

[0010] S4 data analysis, first using historical data to identify the cyclical trends of energy consumption and production to form trend graphs, then using statistical analysis methods to identify the mutual impact of different DERs, and using a machine learning platform to train a prediction model to predict future load and renewable energy output;

[0011] S5. Preliminary strategy formulation: Based on the data analysis results, a preliminary control strategy is formulated. The control strategy includes which resource should be used first and how to dispatch the energy storage system in a specific time period.

[0012] S6. Build a simulation environment. Select power system simulation software to build a simulation model of power supply and demand according to the actual situation, test the initially formulated control strategy in the simulation environment, observe its impact on the balance of power supply and demand, and record the stability, economic benefits and emission reduction effect indicators of the system during the simulation process;

[0013] S7, test strategy, first change variables, test the strategy, variables include market prices and weather conditions, analyze the results of simulation test, identify existing problems, adjust the control strategy based on the test results, retrain the prediction model and change the decision rules, and then conduct simulation test again to verify the effectiveness of the adjusted strategy;

[0014] S8, strategy implementation, applying the optimized control strategy to the actual VPP operation. VPP is a virtual power plant. A monitoring system is established to track the control effect in real time and continuously adjust the control strategy according to the actual operation situation.

[0015] S9. Continuously optimize, regularly evaluate the operating results of the entire system, and optimize the control methods.

[0016] As a preferred energy storage control method for a virtual power plant of the present invention, the metering sensor and metering equipment in S1 are one or more of an electric meter, an inverter, a current transformer, a battery management system, a battery status indicator, a battery temperature sensor, a smart socket, a load controller and a multi-function meter.

[0017] As a preferred energy storage control method for a virtual power plant of the present invention, when cleaning the data in S2, a Python script is written, and the Pandas library is used to clean the data. When standardizing the data, the Z-score standardization method is used to standardize the data.

[0018] As a preferred energy storage control method for a virtual power plant of the present invention, when establishing a data warehouse in S3, Apache Hadoop is used to build the data warehouse, and when integrating and unifying the data, a data fusion algorithm is used for integration.

[0019] As a preferred energy storage control method for a virtual power plant of the present invention, when generating a trend graph in S4, the Matplotlib library of Python is used to draw a trend graph of historical data.

[0020] As a preferred energy storage control method for a virtual power plant of the present invention, in S5, when formulating the preliminary control strategy of the system, a multi-objective optimization algorithm is used. The multi-objective optimization algorithm defines multiple objective functions, including cost minimization and maximization of renewable energy utilization, and then assigns weights to each objective according to actual needs to form a comprehensive evaluation function, and uses the multi-objective optimization algorithm to solve the optimal solution of the comprehensive evaluation function.

[0021] As a preferred energy storage control method for a virtual power plant of the present invention, in S6, MATLAB / Simulink is used to build a VPP simulation model, and the VPP simulation model includes photovoltaics, an energy storage system and a controllable load.

[0022] As a preferred energy storage control method for a virtual power plant of the present invention, when analyzing the results of the simulation test in S7, the power supply and demand balance, economic benefits and emission reduction effects are analyzed.

[0023] As a preferred energy storage control method for a virtual power plant of the present invention, the control effect in S8 includes power supply and demand matching conditions, economic indicators and environmental benefit data.

[0024] A system for an energy storage control method for a virtual power plant, comprising:

[0025] A data acquisition module, wherein the data acquisition module includes a metering sensor and a metering device;

[0026] Edge computing and data preprocessing module, which includes edge computing equipment, data cleaning tools and standardization processing tools;

[0027] A data integration module, the data integration module includes a data storage module and a data fusion module;

[0028] A data analysis module, which includes a big data processing framework, data analysis software, and a machine learning platform;

[0029] Regulatory strategy formulation module;

[0030] Simulation and testing module;

[0031] Strategy optimization and adjustment module;

[0032] Implementation and monitoring module;

[0033] Continuously optimize modules.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. The present invention removes outliers and fills missing values ​​through data cleaning, thereby reducing analysis errors caused by data quality problems. It uses standardized methods such as Z-score to process data so that data from different sources can be compared under the same standard, thereby improving the consistency and reliability of data analysis.

[0036] 2. The present invention establishes a data warehouse to centrally store and manage historical and real-time data, facilitates rapid retrieval and analysis, and uses data fusion algorithms to integrate data from different devices, thereby enhancing the overall coordination of the system and making information exchange between components smoother.

[0037] 3. The present invention uses historical data to identify trends in energy production and consumption, providing a solid foundation for predictive modeling. By training the predictive model, the accuracy of future load and renewable energy output forecasts is improved, providing a scientific basis for formulating strategies.

[0038] 4. The present invention adopts a multi-objective optimization algorithm, comprehensively considers multiple objectives such as cost minimization and maximization of renewable energy utilization, realizes the rational allocation of resources, adjusts the weight of each objective according to actual needs, and forms a flexible and effective control strategy.

[0039] 5. The present invention can discover potential problems and solve them in advance by testing the control strategy in a safe simulation environment, avoiding the risks that may be brought about by direct application to actual operations. The strategy is adjusted based on the simulation test results to ensure the effectiveness and feasibility of the strategy before implementation. A monitoring system is established to achieve real-time tracking of the control effect, so that managers can respond to changes in a timely manner, improve operational efficiency, and quickly adjust strategies based on monitoring data, shortening the decision-making cycle and enhancing the ability to respond to emergencies.

[0040] 6. The present invention promotes the use of clean energy and is beneficial to environmental protection by optimizing the utilization rate of renewable energy. The continuous optimization process allows the control method to be adjusted as the external environment changes, ensuring the long-term stability and adaptability of the system. The system performance is regularly evaluated and the control technology is improved accordingly, promoting technological progress and continuous improvement of service levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0042] Example 1

[0043] See also Figure 1 , a method for regulating energy storage in a virtual power plant, comprising the following steps:

[0044] S1. Data collection: Metering sensors and metering equipment are installed at DERs to collect data on power generation, energy storage status, and load status. DERs are various distributed energy devices or systems that make up the virtual power plant. Wired or wireless communication networks are configured to connect metering sensors and metering equipment to ensure that data can be transmitted from DERs to the central control system.

[0045] S2. Edge computing deployment: Edge computing devices are installed at data collection points. The central control system transmits the received data to the edge computing devices. The edge computing devices perform preliminary data preprocessing. During preprocessing, data cleaning tools are used to remove outliers and fill in missing values ​​to ensure data quality. The data is standardized so that data from different sources can be compared under the same dimension.

[0046] S3 data integration, first build a data warehouse to store historical data and real-time data, and then use the data fusion platform to integrate data from different sources into a unified data set;

[0047] S4 data analysis, first using historical data to identify the cyclical trends of energy consumption and production to form trend graphs, then using statistical analysis methods to identify the mutual impact of different DERs, and using a machine learning platform to train a prediction model to predict future load and renewable energy output;

[0048] S5. Preliminary strategy formulation: Based on the data analysis results, a preliminary control strategy is formulated. The control strategy includes which resource should be used first and how to dispatch the energy storage system in a specific time period.

[0049] S6. Build a simulation environment. Select power system simulation software to build a simulation model of power supply and demand according to the actual situation, test the initially formulated control strategy in the simulation environment, observe its impact on the balance of power supply and demand, and record the stability, economic benefits and emission reduction effect indicators of the system during the simulation process;

[0050] S7, test strategy, first change variables, test the strategy, variables include market prices and weather conditions, analyze the results of simulation test, identify existing problems, adjust the control strategy based on the test results, retrain the prediction model and change the decision rules, and then conduct simulation test again to verify the effectiveness of the adjusted strategy;

[0051] S8, strategy implementation, applying the optimized control strategy to the actual VPP operation. VPP is a virtual power plant. A monitoring system is established to track the control effect in real time and continuously adjust the control strategy according to the actual operation situation.

[0052] S9. Continuously optimize, regularly evaluate the operating results of the entire system, and optimize the control methods.

[0053] Furthermore, the metering sensor and metering device in S1 are one or more of an electric meter, an inverter, a current transformer, a battery management system, a battery status indicator, a battery temperature sensor, a smart socket, a load controller and a multi-function meter.

[0054] Furthermore, when cleaning the data in S2, a Python script was written and the Pandas library was used for data cleaning. When standardizing the data, the Z-score standardization method was used to standardize the data.

[0055] Furthermore, when establishing a data warehouse in S3, Apache Hadoop is used to build the data warehouse, and when integrating and unifying the data, a data fusion algorithm is used for integration.

[0056] Furthermore, when generating trend graphs in S4, Python's Matplotlib library is used to draw historical data trend graphs.

[0057] Furthermore, in S5, a multi-objective optimization algorithm is used in the preliminary regulation strategy of the system. The multi-objective optimization algorithm defines multiple objective functions, including cost minimization and renewable energy utilization maximization. Then, a weight is assigned to each objective according to actual needs to form a comprehensive evaluation function. The multi-objective optimization algorithm is used to solve the optimal solution of the comprehensive evaluation function.

[0058] Furthermore, MATLAB / Simulink is used in S6 to build a VPP simulation model, which includes photovoltaics, energy storage systems, and controllable loads.

[0059] Furthermore, when analyzing the results of the simulation test in S7, the balance of power supply and demand, economic benefits and emission reduction effects are analyzed.

[0060] Furthermore, the regulatory effects in S8 include electricity supply and demand matching, economic indicators and environmental benefit data.

[0061] By removing outliers and filling missing values ​​through data cleaning, analytical errors caused by data quality issues are reduced. Standardization methods such as Z-score are used to process data so that data from different sources can be compared under the same standard, improving the consistency and reliability of data analysis.

[0062] By establishing a data warehouse to centrally store and manage historical and real-time data, it is easy to quickly retrieve and analyze. By using data fusion algorithms to integrate data from different devices, the overall coordination of the system is enhanced, making information exchange between components smoother.

[0063] By using historical data to identify trends in energy production and consumption, a solid foundation is provided for predictive modeling. By training the predictive model, the accuracy of forecasts for future load and renewable energy output is improved, providing a scientific basis for formulating strategies.

[0064] By adopting a multi-objective optimization algorithm and comprehensively considering multiple goals such as cost minimization and maximization of renewable energy utilization, a reasonable allocation of resources is achieved, and the weight of each goal is adjusted according to actual needs to form a flexible and effective control strategy.

[0065] Testing control strategies in a safe simulation environment can identify potential problems and resolve them in advance, avoiding the risks that may be brought about by direct application to actual operations. Strategies can be adjusted based on simulation test results to ensure the effectiveness and feasibility of strategies before implementation. A monitoring system can be established to achieve real-time tracking of control effects, allowing managers to respond to changes in a timely manner, improve operational efficiency, and quickly adjust strategies based on monitoring data, shortening the decision-making cycle and enhancing the ability to respond to emergencies.

[0066] By optimizing the utilization rate of renewable energy, the use of clean energy is promoted, which is beneficial to environmental protection. The continuous optimization process allows the control method to be adjusted as the external environment changes, ensuring the long-term stability and adaptability of the system. The system performance is regularly evaluated and the control technology is improved accordingly, which promotes technological progress and continuous improvement of service levels.

[0067] Example 2

[0068] A system for an energy storage control method for a virtual power plant, comprising:

[0069] A data acquisition module, wherein the data acquisition module includes a metering sensor and a metering device;

[0070] Edge computing and data preprocessing module, which includes edge computing equipment, data cleaning tools and standardization processing tools;

[0071] A data integration module, the data integration module includes a data storage module and a data fusion module;

[0072] A data analysis module, which includes a big data processing framework, data analysis software, and a machine learning platform;

[0073] Regulatory strategy formulation module;

[0074] Simulation and testing module;

[0075] Strategy optimization and adjustment module;

[0076] Implementation and monitoring module;

[0077] Continuously optimize modules.

[0078] The system's workflow is as follows:

[0079] According to the metering requirements, metering equipment such as electricity meters, inverters, current transformers, battery management systems (BMS), battery status indicators, battery temperature sensors, smart sockets, load controllers and multi-function meters are installed at DERs. Wired or wireless communication networks are configured to ensure that data can be transmitted from DERs to the central control system. The central control system receives the data transmitted from DERs and transmits it to the edge computing device. The edge computing device cleans the data and standardizes the data to ensure data quality. The pre-processed data is stored in the data warehouse. Data fusion algorithms are used to integrate data from different sources into a unified data set. Historical data is used to identify energy consumption and production cycles. The paper analyzes the trend of the power supply and demand of the power plant and draws a trend chart. The paper identifies the mutual influence between different DERs through statistical analysis methods. The paper uses a machine learning platform to train the prediction model to predict future load and renewable energy output. The paper defines multiple objective functions and assigns weights to each objective according to actual needs to form a comprehensive evaluation function. The paper uses a multi-objective optimization algorithm to solve the optimal solution of the comprehensive evaluation function. The paper formulates a preliminary control strategy. The paper builds a VPP simulation model including photovoltaic, energy storage system and controllable load. The paper tests the preliminary control strategy in the simulation environment to observe its impact on the balance of power supply and demand, and records the indicators reflecting the stability, economic benefits and emission reduction effect of the system. The paper tests the strategy by changing variables such as market prices and weather conditions. The paper analyzes the results of the simulation test, identifies the existing problems, adjusts the control strategy based on the test results, retrains the prediction model, changes the decision rules, verifies the effectiveness of the adjusted strategy, applies the optimized control strategy to the actual VPP operation, establishes a monitoring system, tracks the control effect in real time, including the power supply and demand matching, economic indicators and environmental benefit data, continuously adjusts the control strategy according to the actual operation, regularly evaluates the operation effect of the entire system, optimizes the control method according to the evaluation results, and forms a closed-loop optimization mechanism.

[0080] By installing metering equipment at distributed energy resources, energy production and consumption can be monitored in real time to ensure the accuracy and timeliness of data. The configured wired or wireless communication network ensures that data can be transmitted from DERs to the central control system and further transmitted to the edge computing device to achieve rapid data processing and analysis. The edge computing device performs data cleaning and standardization processing to ensure data quality and provide a reliable basis for subsequent analysis. Data from different sources are integrated using data fusion algorithms. The prediction model is trained through statistical analysis and machine learning platforms to more accurately predict energy load and renewable energy output. Multiple objective functions are defined and weights are assigned. Through multi-objective optimization algorithm solution, a control strategy that comprehensively considers multiple factors can be formulated. The control strategy is tested in the simulation model to observe its impact on the balance of power supply and demand, and the strategy is adjusted according to the simulation test results to ensure the effectiveness and adaptability of the strategy. The optimized control strategy is applied to the actual virtual power plant (VPP) operation, and a monitoring system is established to track the control effect in real time to ensure the stability and economic benefits of the system. The control strategy is continuously adjusted according to the actual operation situation, and the operation effect of the entire system is regularly evaluated to form a closed-loop optimization mechanism to continuously improve the performance and benefits of the system.

[0081] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for regulating energy storage in a virtual power plant The method for regulating energy storage in a virtual power plant is characterized in that: The following steps are involved: S1. Data collection: Metering sensors and metering equipment are installed at DERs to collect data on power generation, energy storage status, and load status. DERs are various distributed energy devices or systems that make up the virtual power plant. Wired or wireless communication networks are configured to connect metering sensors and metering equipment to ensure that data can be transmitted from DERs to the central control system. S2. Edge computing deployment: Edge computing devices are installed at data collection points. The central control system transmits the received data to the edge computing devices. The edge computing devices perform preliminary data preprocessing. During preprocessing, data cleaning tools are used to remove outliers and fill in missing values ​​to ensure data quality. The data is standardized so that data from different sources can be compared under the same dimension. S3, data integration, first establish a data warehouse to store historical data and real-time data, and then use the data fusion platform to integrate data from different sources into a unified data set; S4. Data analysis: First, use historical data to identify the cyclical trends of energy consumption and production to form trend graphs. Then, use statistical analysis methods to identify the mutual influence between different DERs, and use machine learning platforms to train prediction models to predict future loads and renewable energy output. S5. Preliminary strategy formulation: Based on the data analysis results, a preliminary control strategy is formulated. The control strategy includes which resource should be used first and how to dispatch the energy storage system in a specific time period. S6. Build a simulation environment. Select power system simulation software to build a simulation model of power supply and demand according to the actual situation, test the initially formulated control strategy in the simulation environment, observe its impact on the balance of power supply and demand, and record the stability, economic benefits and emission reduction effect indicators of the system during the simulation process; S7, test strategy, first change variables, test the strategy, variables include market prices and weather conditions, analyze the results of simulation test, identify existing problems, adjust the control strategy based on the test results, retrain the prediction model and change the decision rules, and then conduct simulation test again to verify the effectiveness of the adjusted strategy; S8, strategy implementation, applying the optimized control strategy to the actual VPP operation. VPP is a virtual power plant. A monitoring system is established to track the control effect in real time and continuously adjust the control strategy according to the actual operation situation. S9. Continuously optimize, regularly evaluate the operating results of the entire system, and optimize the control methods.

2. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: The metering sensor and metering device in S1 are one or more of an electric meter, an inverter, a current transformer, a battery management system, a battery status indicator, a battery temperature sensor, a smart socket, a load controller and a multi-function meter.

3. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: When cleaning the data in S2, a Python script is written and the Pandas library is used to clean the data. When standardizing the data, the Z-score standardization method is used to standardize the data.

4. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: When establishing the data warehouse in the S3, Apache Hadoop is used to build the data warehouse, and when integrating and unifying the data, a data fusion algorithm is used for integration.

5. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: When generating a trend graph in S4, the Matplotlib library of Python is used to draw a trend graph of historical data.

6. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: In the S5, a multi-objective optimization algorithm is used in the preliminary regulation strategy of the system. The multi-objective optimization algorithm defines multiple objective functions, including cost minimization and renewable energy utilization maximization. Then, weights are assigned to each objective according to actual needs to form a comprehensive evaluation function. The multi-objective optimization algorithm is used to solve the optimal solution of the comprehensive evaluation function.

7. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: In S6, MATLAB / Simulink is used to construct a VPP simulation model, which includes photovoltaics, energy storage systems and controllable loads.

8. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: When analyzing the results of the simulation test in S7, the power supply and demand balance, economic benefits and emission reduction effects are analyzed.

9. The energy storage control method for a virtual power plant according to claim 1 is characterized in that: The control effects in S8 include power supply and demand matching, economic indicators and environmental protection benefit data.

10. A system for an energy storage control method for a virtual power plant, applicable to the energy storage control method for a virtual power plant as claimed in any one of claims 1 to 9, characterized in that: include: A data acquisition module, wherein the data acquisition module includes a metering sensor and a metering device; Edge computing and data preprocessing module, which includes edge computing equipment, data cleaning tools and standardization processing tools; A data integration module, the data integration module includes a data storage module and a data fusion module; A data analysis module, which includes a big data processing framework, data analysis software, and a machine learning platform; Regulatory strategy formulation module; Simulation and testing module; Strategy optimization and adjustment module; Implementation and monitoring module; Continuously optimize modules.

Images