Search

CN-120999668-B - Virtual power plant comprehensive energy management system based on optical storage joint debugging control

CN120999668BCN 120999668 BCN120999668 BCN 120999668BCN-120999668-B

Abstract

The invention discloses a virtual power plant comprehensive energy management system based on optical storage joint debugging control, which belongs to the technical field of virtual power plants and comprises a data acquisition module, a data processing module, an analysis prediction module and an energy management module, wherein the data acquisition module is used for acquiring virtual power plant multi-source data, the data processing module is used for processing the virtual power plant multi-source data to determine virtual power plant characteristic data, the analysis prediction module is used for constructing a virtual power plant comprehensive energy management prediction model to analyze the virtual power plant characteristic data to determine virtual power plant comprehensive energy management prediction results, and the energy management module is used for managing virtual power plant comprehensive energy based on optical storage joint debugging control. The virtual power plant comprehensive energy management system solves the problem that the virtual power plant management effect is poor due to the fact that the virtual power plant comprehensive energy cannot be effectively managed in the prior art. The invention can effectively manage the comprehensive energy of the virtual power plant based on the optical storage joint debugging control, can effectively realize the efficient utilization of the energy and the flexible adjustment of the power grid, and can promote the management effect of the virtual power plant.

Inventors

  • YANG JIYONG

Assignees

  • 苏州通合智慧能源有限公司

Dates

Publication Date
20260505
Application Date
20250729

Claims (8)

  1. 1. Virtual power plant integrated energy management system based on optical storage joint debugging control, which is characterized by comprising: the data acquisition module is used for acquiring virtual power plant multi-source data based on the Internet of things equipment; the data processing module is used for processing the multi-source data of the virtual power plant and determining the characteristic data of the virtual power plant; Detecting the virtual power plant multi-source data in real time, identifying abnormal values in the virtual power plant multi-source data, and judging whether the abnormal values in the virtual power plant multi-source data are valuable for virtual power plant comprehensive energy management or not based on the contribution rate; the analysis and prediction module is used for constructing a virtual power plant comprehensive energy management prediction model to analyze the virtual power plant characteristic data and determine a virtual power plant comprehensive energy management prediction result; the energy management module is used for managing the comprehensive energy of the virtual power plant based on the optical storage joint debugging control; judging whether abnormal values in the virtual power plant multi-source data are valuable for virtual power plant comprehensive energy management based on the contribution rate, comprising: Performing value evaluation on the abnormal values in the multi-source data based on the contribution rates of the technical dimension, the economic dimension and the risk dimension, and determining the evaluation value of the abnormal values in the multi-source data on the comprehensive energy management of the virtual power plant; Wherein, the An evaluation value representing the abnormal value for the virtual power plant comprehensive energy management; 、 、 Respectively the weight coefficients; S represents the total number of categories of the data sources; N represents the total number of abnormal values in the data source of the s class; Representing a historical average of data values in the class s data source; Representing a historical standard deviation of data values in a class s data source; representing the sensitivity of the system energy efficiency to the type s data source, namely: ; Representing a cost difference due to the abnormal data; Representing a reference cost reference value; represents a time decay factor, t represents an abnormality occurrence time, Representing the electricity price peak value center point; representing a time decay coefficient; Representing the cascade failure probability of the system; Representing the fault probability of the kth subsystem; k represents a subsystem index; representing abnormality of class s data source During the time, the fault probability of the subsystem k; Representing a risk tolerance threshold; the evaluation value is calculated Comparing with a preset evaluation value threshold value, if the evaluation value If the value is larger than or equal to a preset evaluation value threshold, the abnormal value in the multi-source data is valuable for the comprehensive energy management of the virtual power plant; if the evaluation value is If the abnormal value in the multi-source data is smaller than the preset evaluation value threshold, the abnormal value in the multi-source data has no value for the comprehensive energy management of the virtual power plant; constructing a virtual power plant comprehensive energy management prediction model, and executing the following operations: collecting virtual power plant historical data, and dividing the collected virtual power plant historical data, wherein the virtual power plant historical data is divided into a training set and a testing set; training a deep learning model by adopting a training set based on a deep learning technology, enabling the deep learning model to autonomously learn virtual power plant comprehensive energy management prediction behaviors from the training set, predicting supply and demand conditions of the virtual power plant, and determining a virtual power plant comprehensive energy management prediction model based on deep learning; Inputting the test set into a virtual power plant comprehensive energy management prediction model based on deep learning, testing the virtual power plant comprehensive energy management prediction model based on deep learning based on the test set, evaluating the performance of the virtual power plant comprehensive energy management prediction model based on deep learning, judging whether the virtual power plant comprehensive energy management prediction model based on deep learning can reach the expected effect of virtual power plant comprehensive energy management prediction, and determining a model test evaluation result; And when the virtual power plant comprehensive energy management prediction model based on the deep learning cannot reach the expected effect of the virtual power plant comprehensive energy management prediction, carrying out parameter adjustment and iterative optimization on the virtual power plant comprehensive energy management prediction model based on the deep learning until the virtual power plant comprehensive energy management prediction model based on the deep learning can reach the expected effect of the virtual power plant comprehensive energy management prediction, thereby determining the optimal virtual power plant comprehensive energy management prediction model.
  2. 2. The integrated energy management system for a virtual power plant based on optical joint debugging control according to claim 1, wherein the real-time detection of the virtual power plant multi-source data and the identification of the abnormal value in the virtual power plant multi-source data comprise: Sequencing and aligning the virtual power plant multi-source data based on a time sequence to obtain aligned virtual power plant multi-source data; Taking one data source data in the aligned virtual power plant multi-source data as data to be identified; Obtaining the extreme value of each data point in the data to be identified in the neighborhood range; dividing regions of the data to be identified based on continuous data points with the same extremum, and determining a plurality of time sequence fragments; respectively calculating correlation coefficients between any two time sequence segments; comparing the correlation coefficient with a preset correlation coefficient threshold value, and forming a time sequence fragment group by the time sequence fragments when the correlation coefficient is larger than or equal to the preset correlation coefficient threshold value, so as to obtain a plurality of time sequence fragment groups; Respectively calculating entropy and average value of data values in each time sequence segment group; determining an evaluation coefficient of each time sequence segment based on the entropy and the average value; Calculating the average value of the evaluation coefficients of all the time sequence fragments in each time sequence fragment group, and determining the average difference of the evaluation coefficients of each time sequence fragment based on the average value; taking the time sequence fragments with the average deviation greater than or equal to a preset average deviation threshold value as abnormal time sequence fragments; Taking any abnormal time sequence segment as a first abnormal time sequence segment, and taking a plurality of normal time sequence segments in a time sequence segment group where the first abnormal time sequence segment is positioned as second time sequence segments; calculating the data mean value of the position data point corresponding to each data point in the plurality of second time sequence fragments; Performing difference operation on the data points in the first abnormal time sequence segment and the data mean value to obtain a deviation value of each data point in the first abnormal time sequence segment; taking the data point when the deviation value is greater than or equal to a preset deviation threshold value as a different data point; Traversing all abnormal time sequence segments to obtain a plurality of abnormal data points; and traversing all data source data in the virtual power plant multi-source data, and determining abnormal values corresponding to abnormal data points in the virtual power plant multi-source data.
  3. 3. The virtual power plant integrated energy management system based on optical storage joint debugging control according to claim 2, wherein the virtual power plant integrated energy management prediction model is constructed to analyze the virtual power plant characteristic data, and the following operations are performed: Deploying the virtual power plant comprehensive energy management prediction model, and deploying the virtual power plant comprehensive energy management prediction model in a virtual power plant comprehensive energy management prediction environment based on optical storage joint debugging control; And inputting the virtual power plant characteristic data into a virtual power plant comprehensive energy management prediction model, analyzing the virtual power plant characteristic data according to the virtual power plant comprehensive energy management prediction model, identifying the virtual power plant comprehensive energy management condition, predicting the virtual power plant supply and demand condition, and determining the virtual power plant comprehensive energy management prediction result.
  4. 4. The virtual power plant integrated energy management system based on optical storage joint debugging control according to claim 3, wherein the fluctuation of photovoltaic power generation is smoothed by adopting energy storage equipment according to the virtual power plant integrated energy management prediction result, peak clipping and valley filling and demand response are realized, and the virtual power plant integrated energy is effectively managed based on the optical storage joint debugging control; when the distributed photovoltaic power generation quantity meets the power load requirement, the residual electric energy exists, and the residual electric energy is stored by adopting energy storage equipment; when the distributed photovoltaic power generation capacity cannot meet the power load demand, the energy storage equipment is used for releasing the stored electric energy to meet the power load demand.
  5. 5. The integrated energy management system of a virtual power plant based on optical joint debugging control according to claim 4, wherein the virtual power plant multi-source data is collected, and the following operations are performed: Real-time monitoring and collecting real-time power generation data, environment data and equipment state data of a power generation side based on Internet of things equipment to obtain distributed photovoltaic power generation data; Based on the internet of things equipment, monitoring and acquiring the charge and discharge state of the energy storage equipment and the data of the battery pack in real time to acquire the data of the energy storage equipment; Real-time monitoring and acquisition are carried out on the real-time electricity utilization data and the adjustable capacity data of the electricity utilization side based on the Internet of things equipment, and controllable load data are obtained; and determining the virtual power plant multi-source data according to the distributed photovoltaic power generation data, the energy storage equipment data and the controllable load data.
  6. 6. The virtual power plant integrated energy management system based on optical storage joint debugging control of claim 5, wherein the real-time power generation data comprises photovoltaic array output power, direct-current side voltage and current, alternating-current side voltage, current and frequency, inverter efficiency and power factor; the environmental data comprise illumination intensity, environmental temperature, photovoltaic module temperature, wind speed and wind direction; the equipment state data comprise an inverter running state, a photovoltaic group running state and a grid-connected point voltage state; The charge and discharge state comprises the current charge and discharge power, charge and discharge current, charge and discharge efficiency, the residual battery capacity, the battery health state and the discharge depth; the battery pack data comprises single battery voltage, total battery voltage, battery temperature and internal resistance; the real-time electricity data comprise load power, voltage, current, power factor and load curve; the adjustability data includes maximum curtailable power, minimum operating power, interruptible time, and resume time.
  7. 7. The integrated energy management system of a virtual power plant based on optical joint debugging control according to claim 6, wherein the processing of the virtual power plant multi-source data performs the following operations: The virtual power plant multi-source data are cleaned, noise data which are valuable to virtual power plant comprehensive energy management in the virtual power plant multi-source data are removed, and interference of noise to the virtual power plant comprehensive energy management is reduced; Detecting the virtual power plant multi-source data in real time, identifying abnormal values in the virtual power plant multi-source data, and processing the abnormal values in the virtual power plant multi-source data; The method comprises the steps of evaluating abnormal values in virtual power plant multi-source data, and judging whether the abnormal values in the virtual power plant multi-source data are valuable for virtual power plant comprehensive energy management or not based on contribution rates; When the abnormal value in the virtual power plant multi-source data is valuable for the virtual power plant comprehensive energy management, correcting the abnormal value in the virtual power plant multi-source data; And when the abnormal value in the virtual power plant multi-source data is not valuable for the virtual power plant comprehensive energy management, deleting the abnormal value in the virtual power plant multi-source data.
  8. 8. The integrated energy management system for a virtual power plant based on optical joint debugging control according to claim 7, wherein the processing of the virtual power plant multi-source data is performed by: Normalizing the virtual power plant multi-source data to convert the virtual power plant multi-source data into a unified data format, and removing dimension differences in the virtual power plant multi-source data to form standardized virtual power plant multi-source data; Integrating the virtual power plant multi-source data, integrating the virtual power plant multi-source data with different sources into a unified data view, and storing and backing up the integrated virtual power plant multi-source data; And extracting the characteristics of the virtual power plant multi-source data, extracting the characteristic vectors related to the virtual power plant comprehensive energy management from the virtual power plant multi-source data, and carrying out weighted fusion on the extracted characteristic vectors to form the virtual power plant characteristic data.

Description

Virtual power plant comprehensive energy management system based on optical storage joint debugging control Technical Field The invention relates to the technical field of virtual power plants, in particular to a virtual power plant comprehensive energy management system based on optical storage joint debugging control. Background The virtual power plant is an intelligent energy system which integrates distributed energy resources (such as photovoltaic power generation, energy storage equipment, adjustable load and the like) through advanced information communication technology and digital means and realizes centralized optimization management. The intelligent scheduling method has the core functions of resource aggregation, integration of distributed power supply, energy storage and controllable load to form a virtual power plant, intelligent scheduling, dynamic adjustment of energy production and consumption by utilizing artificial intelligence, big data analysis and real-time monitoring technology, and market participation, wherein economic benefits are obtained by participating in electric power market transaction. The prior art cannot effectively manage comprehensive energy of the virtual power plant based on optical storage joint debugging control, cannot effectively realize efficient utilization of energy and flexible adjustment of a power grid, and causes poor management effect of the virtual power plant. Disclosure of Invention The invention aims to provide a virtual power plant comprehensive energy management system based on optical storage joint debugging control, which can effectively manage virtual power plant comprehensive energy based on optical storage joint debugging control, can effectively realize efficient utilization of energy and flexible regulation of a power grid, can promote virtual power plant management effect, and solves the problems in the background technology. In order to achieve the above purpose, the present invention provides the following technical solutions: virtual power plant integrated energy management system based on optical storage joint debugging control includes: the data acquisition module is used for acquiring virtual power plant multi-source data based on the Internet of things equipment; the data processing module is used for processing the multi-source data of the virtual power plant and determining the characteristic data of the virtual power plant; the analysis and prediction module is used for constructing a virtual power plant comprehensive energy management prediction model to analyze the virtual power plant characteristic data and determine a virtual power plant comprehensive energy management prediction result; and the energy management module is used for managing the comprehensive energy of the virtual power plant based on the optical storage joint debugging control. Preferably, the virtual power plant characteristic data is analyzed by constructing a virtual power plant comprehensive energy management prediction model, and the following operations are executed: Deploying the virtual power plant comprehensive energy management prediction model, and deploying the virtual power plant comprehensive energy management prediction model in a virtual power plant comprehensive energy management prediction environment based on optical storage joint debugging control; And inputting the virtual power plant characteristic data into a virtual power plant comprehensive energy management prediction model, analyzing the virtual power plant characteristic data according to the virtual power plant comprehensive energy management prediction model, identifying the virtual power plant comprehensive energy management condition, predicting the virtual power plant supply and demand condition, and determining the virtual power plant comprehensive energy management prediction result. Preferably, according to the comprehensive energy management prediction result of the virtual power plant, the fluctuation of the photovoltaic power generation is smoothed by adopting the energy storage equipment, peak clipping, valley filling and demand response are realized, and the comprehensive energy of the virtual power plant is effectively managed based on the optical storage joint debugging control; when the distributed photovoltaic power generation quantity meets the power load requirement, the residual electric energy exists, and the residual electric energy is stored by adopting energy storage equipment; when the distributed photovoltaic power generation capacity cannot meet the power load demand, the energy storage equipment is used for releasing the stored electric energy to meet the power load demand. Preferably, virtual power plant multisource data is collected, and the following operations are performed: Real-time monitoring and collecting real-time power generation data, environment data and equipment state data of a power generation side based on Internet of things equipment to ob