Virtual power plant energy management method and system based on air conditioning load

Abstract

The invention discloses a virtual power plant energy management method and system based on an air conditioning load, and relates to the field of energy management, and the management method comprises the steps: obtaining power consumption data and integrated operation information data in a virtual power plant management range, carrying out the statistical division of the operation information data according to an operation rule and a potential distribution model, and carrying out the statistical division of the operation information data; generating a batch monitoring management set; constructing a virtual power plant electric energy transaction framework based on the batch monitoring management set, analyzing an electric power balance relationship between an electric quantity supply and demand difference and a virtual power plant, and outputting an energy scheduling result in a virtual power plant management range; and the air conditioner load data serve as demand response resources, the energy dispatching result is optimized, an air conditioner control issuing instruction is generated, the air conditioner control issuing instruction is recorded and decomposed, and the control state of the air conditioner is output. Through optimal scheduling of the air conditioner load, the demand of the air conditioner load can be reduced in the peak period of power demand, so that the load pressure of the system is effectively reduced, and excessive energy consumption is avoided.

Description

Technical Field

[0001] The present invention relates to the field of energy management, and particularly to a virtual power plant energy management method and system based on air-conditioning load. Background Art

[0002] A virtual power plant (VPP) is a system that, through information technology means, intelligently schedules and manages dispersed and heterogeneous distributed energy resources (such as wind energy, solar energy, energy storage, electric vehicles, electric water heaters, air conditioners, etc.), thereby showing the ability of large-scale power production and consumption in the form of "virtual" in the power market. It realizes the balance between supply and demand and improves the stability and economy of the power system through means such as demand response, energy storage, and predictive scheduling.

[0003] Energy management in a virtual power plant (VPP) is one of its core capabilities, which determines whether the virtual power plant can achieve efficient scheduling, load balancing, economic operation, and the ability to participate in market transactions. The so-called virtual power plant energy management refers to the whole process of intelligently monitoring, optimizing scheduling, and dynamically coordinating the distributed power sources, energy storage devices, load resources, etc. aggregated by the virtual power plant.

[0004] However, in the prior art, the virtual power plant often ignores the load changes of air conditioners within the management scope during the process of realizing energy scheduling management. When a large number of air-conditioning devices participate in energy scheduling, the usage patterns of air conditioners fluctuate greatly, resulting in inaccurate energy scheduling results, such as excessive energy consumption or insufficient energy for other electrical equipment. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes a virtual power plant energy management method and system based on air-conditioning load to achieve the purpose of avoiding excessive energy consumption.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] In the first aspect, the present invention provides a virtual power plant energy management method based on air-conditioning load, and the management method includes:

[0008] Obtain the electricity consumption data within the management scope of the virtual power plant, integrate the operation information data, and perform unified classification on the operation information data according to the operation rules and potential allocation models to generate a batch monitoring and management set;

[0009] Based on the batch monitoring and management set, construct a virtual power plant electricity energy trading framework, analyze the difference between electricity supply and demand and the power balance relationship between virtual power plants, and output the energy scheduling result within the management scope of the virtual power plant;

[0010] Taking the air-conditioning load data as demand response resources, optimizing the energy dispatch result to generate the air-conditioning control issuance instruction, and recording and decomposing the air-conditioning control issuance instruction to output the control status of the air conditioner.

[0011] Preferably, statistically classifying the operation information data according to the operation rules and the potential allocation model, and generating a batch monitoring and management set including:

[0012] Analyzing the joint probability of the Naive Bayes model based on the hyperparameters of the tags and values involved in the operation information data, optimizing the Naive Bayes model based on the analysis results, and judging the correlation between tags and the correlation between values;

[0013] Analyzing the weights of tags and values respectively based on the correlation between tags and the correlation between values, outputting the main features of the operation information data according to the weight results, and combining with the operation rules to determine the classification category tags of the operation information data, where the classification category tags include industry classification, regional classification, and user classification;

[0014] Analyzing the joint distribution dependence between industry classification, regional classification, and user classification and the main features, clustering the dependence, and training the Naive Bayes model based on the clustering results to obtain the potential allocation model;

[0015] Generating classification models themed on industry classification, regional classification, and user classification respectively based on the training results, and using the classification models to statistically analyze the operation information data according to industry, region, and user to obtain the batch monitoring and management set.

[0016] Preferably, constructing a virtual power plant electric energy trading framework based on the batch monitoring and management set, analyzing the difference between power supply and demand and the power balance relationship between virtual power plants, and outputting the energy dispatch result within the management scope of the virtual power plant including:

[0017] Obtaining the historical power generation of each energy within the management scope of the virtual power plant, predicting the renewable energy generation combined with environmental factor data, and constructing a virtual power plant electric energy trading framework based on the renewable energy generation and electricity price;

[0018] Batch predicting the load fluctuation trend managed by the virtual power plant based on the batch monitoring and management set, and analyzing the difference between power supply and demand during the management process of the virtual power plant according to the renewable energy generation and the load fluctuation trend;

[0019] Evaluating the difference between power supply and demand and the power balance relationship between virtual power plants based on the virtual power plant electric energy trading framework, and determining the dispatching mechanism of the virtual power plant electric energy trading by analyzing the difference according to the power balance relationship.

[0020] Preferably, obtaining the historical power generation of each energy within the management scope of the virtual power plant, predicting the renewable energy generation combined with environmental factor data, and constructing a virtual power plant electric energy trading framework including:

[0021] Extract the historical power generation of each energy within the management scope of the virtual power plant based on grid monitoring devices, and obtain the environmental factor data within the management scope of the virtual power plant according to the demand during the prediction period;

[0022] Define the predicted consecutive days and predicted consecutive weeks according to the historical power generation, analyze the number of power generation change periods and power generation data within the predicted consecutive days, and calculate the daily power generation change factor and the daily average change factor;

[0023] Analyze the number of power generation change periods and power generation data within the predicted consecutive weeks, calculate the weekly power generation change factor and the weekly average change factor, and combine the calculation results with the environmental factor data to predict the renewable energy generation;

[0024] Based on the prediction results, design the virtual power plant electric energy trading rules, construct the virtual power plant electric energy trading framework, and analyze the energy cost within the management scope of the virtual power plant under the condition of predicted renewable energy generation.

[0025] Preferably, combining the calculation results with the environmental factor data to predict the renewable energy generation includes:

[0026] Judge the simulated power generation during the prediction period according to the daily average change factor and the weekly average change factor, and determine the influencing factors affecting the energy power generation result based on the environmental factor data;

[0027] Based on the influencing factors, respectively consider the ratios of the daily average change factor and the weekly average change factor in the simulated power generation, and analyze the sum of squared errors of the simulated power generation according to the ratio results;

[0028] Determine the comprehensive weight factor by minimizing the sum of squared errors, determine the selection results of the predicted consecutive days and the predicted consecutive weeks based on the comprehensive weight factor, and obtain the predicted renewable energy generation according to the selection results;

[0029] Obtain the day-ahead electricity price data, formulate the virtual power plant's quotation rules in combination with the predicted renewable energy generation, and select the trading market according to the quotation results to provide a basis for the construction of the virtual power plant electric energy trading framework.

[0030] Preferably, the calculation formula for the predicted renewable energy generation is:

[0031] ;

[0032] In the formula, E a,b represents the predicted renewable energy generation of energy in the b time period during the a - th acquisition process, ω represents the comprehensive weight factor, R represents the total number of predicted consecutive days, G r,b represents the power generation in the b time period of the r - th predicted consecutive day, represents the daily average power generation of the r-th predicted consecutive day, D represents the total number of predicted consecutive weeks, D d,b represents the power generation of the b time period in the d-th predicted consecutive week, represents the weekly average power generation of the d-th predicted consecutive week, represents the average power generation within the scope of virtual power plant management.

[0033] Preferably, batch predict the load fluctuation trend of virtual power plant management based on the batch monitoring management set, and analyze the power supply-demand balance in the process of virtual power plant management according to the renewable energy generation and the load fluctuation trend, including:

[0034] Based on the batch monitoring management set, determine the load simulation targets and load evaluation indicators corresponding to industry classification, regional classification, and user classification respectively, and set the distribution and parameters of random sampling to conduct random sampling;

[0035] According to the random sampling results and statistical simulation techniques, construct a difference matrix to generate random operation information samples, and use the attribute theory based on information theory to reduce the random operation information samples to construct a trend prediction model;

[0036] Based on the trend prediction model, predict the load fluctuation trends corresponding to industry classification, regional classification, and user classification respectively, and analyze the electricity demand within the scope of virtual power grid management according to the load fluctuation trends;

[0037] Compare the difference between the predicted results of electricity demand and renewable energy generation, determine the energy profit and loss status in the process of virtual power plant management according to the comparison results, and analyze the power supply-demand balance based on the energy profit and loss status.

[0038] Preferably, according to the random sampling results and statistical simulation techniques, construct a difference matrix to generate random operation information samples, and use the attribute theory based on information theory to reduce the random operation information samples to construct a trend prediction model, including:

[0039] Based on the load evaluation indicators, simulate and judge the reliability of the random sampling results, optimize and adjust the random sampling results according to the reliability results until the reliability meets the load simulation target to obtain random operation information samples;

[0040] Map the random operation information samples to the attribute space to construct a difference matrix, traverse and search the data in the difference matrix until an empty set is encountered and then stop the search, and obtain a reduced triangular matrix according to the search results;

[0041] Analyze the original sub-attribute assignments of the random operation information samples in the reduced triangular matrix, and verify them with the original sub-attribute assignments corresponding to the difference matrix, and judge the reduction accuracy of the reduced triangular matrix according to the verification results;

[0042] Take the random operation information sample corresponding to the reduced triangular matrix that meets the reduction requirements as the data set, and use the data set as the input and the load fluctuation as the output to construct a trend prediction model in combination with the Monte Carlo algorithm.

[0043] Preferably, use the air-conditioning load data as the demand response resource, optimize the energy dispatch result to generate the air-conditioning control instruction for distribution, and record and decompose the air-conditioning control instruction for distribution, and the output air-conditioning control status includes:

[0044] Obtain the power load data of various air conditioners within the management scope of the virtual power plant as the demand response resource, judge the demand response volume of the air-conditioning power load data during the prediction period, and adjust the energy dispatch result based on the demand response volume;

[0045] Determine the control strategy of various air conditioners according to the adjusted energy dispatch result, determine the air-conditioning control instruction for distribution, and send the air-conditioning control instruction for distribution to the device gateway layer through the message transmission protocol;

[0046] The device gateway layer receives and analyzes the air-conditioning control instruction for distribution, obtains the power information required for each air conditioner to adjust the control status of the air conditioner, and evaluates the effect of the execution status of the air conditioners participating in the control.

[0047] In a second aspect, the present invention also provides a virtual power plant energy management system based on air-conditioning load. The management system includes:

[0048] A monitoring and management set generation module, which is used to obtain the power consumption data within the management scope of the virtual power plant, integrate the operation information data, and uniformly divide the operation information data according to the operation rules and the potential allocation model to generate a batch monitoring and management set;

[0049] An energy dispatch result output module, which is used to construct a virtual power plant electricity energy trading framework based on the batch monitoring and management set, analyze the difference between electricity supply and demand and the power balance relationship between virtual power plants, and output the energy dispatch result within the management scope of the virtual power plant;

[0050] An air-conditioning control status output module, which is used to use the air-conditioning load data as the demand response resource, optimize the energy dispatch result to generate the air-conditioning control instruction for distribution, and record and decompose the air-conditioning control instruction for distribution, and output the air-conditioning control status.

[0051] The beneficial effects of the present invention are:

[0052] 1. In the present invention, by using air-conditioning load data as demand response resources, the virtual power plant can flexibly adjust the air-conditioning load according to the fluctuations of power demand, thereby balancing the power supply and demand of the virtual power plant, greatly improving the adaptability of the power system during load fluctuations. At the same time, by using the electricity energy trading framework to analyze the difference between supply and demand, it is possible to predict in advance the load changes of the power system. Through the optimal scheduling of the air-conditioning load, the demand for the air-conditioning load can be reduced during the peak power demand period, thereby effectively reducing the system load pressure and avoiding excessive energy consumption.

[0053] 2. In the present invention, by combining historical power generation with environmental factors such as weather, temperature, and season, the short-term and medium-term renewable energy output can be predicted, which can address in advance the problem of strong volatility of renewable energy, avoid the abandonment of electricity due to over-generation or passive emergency due to insufficient power generation. At the same time, after the difference between power supply and demand is evaluated, the virtual power plant can adopt different scheduling mechanisms according to the balance relationship, realizing the automation and intelligence of energy scheduling and enhancing the economic benefits of the virtual power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention.

[0055] Figure 1 is a flowchart of a virtual power plant energy management method based on air-conditioning load according to an embodiment of the present invention;

[0056] Figure 2 is a schematic block diagram of a virtual power plant energy management system based on air-conditioning load according to an embodiment of the present invention.

[0057] In the figure:

[0058] 1. Monitoring and management set generation module; 2. Energy scheduling result output module; 3. Air-conditioning control status output module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0060] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0061] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular forms are also intended to include the plural forms. In addition, it should be understood that the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0062] In the case of no conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0063] Please refer to Figure 1 , the present invention provides a virtual power plant energy management method based on air conditioning load, and the management method includes:

[0064] Step S1, obtain the electricity consumption data within the management scope of the virtual power plant, integrate the operation information data, and make a unified classification of the operation information data according to the operation rules and the potential allocation model to generate a batch monitoring management set.

[0065] In one embodiment, in the process of statistically classifying operation information data according to operation rules and a potential allocation model to generate a batch monitoring management set, the joint probability of a Naive Bayes model can be analyzed based on the hyperparameters of the tags and values involved in the operation information data. Based on the analysis results, the Naive Bayes model is optimized to judge the correlation between tags and the correlation between values. Based on the correlation between tags and the correlation between values, the weights of tags and values are respectively analyzed. According to the weight results, the main features of the operation information data are output, and combined with the operation rules to determine the classification category labels of the operation information data. The classification category labels include industry classification, regional classification, and user classification. Analyze the joint distribution dependence between industry classification, regional classification, and user classification and the main features, and cluster the dependence. Based on the clustering results, train the Naive Bayes model to obtain a potential allocation model. Based on the training results, classification models with industry classification, regional classification, and user classification as themes are respectively generated. The classification models are used to statistically analyze the operation information data according to industry, region, and user to obtain a batch monitoring management set. At the same time, in the process of training the Naive Bayes model to prevent the model from overfitting, a cross-validation mechanism is introduced during the training process, and the model structure is simplified. At the same time, regularization strategies (such as L1 / L2) and early stopping methods (Early Stopping) are used to control the number of training rounds. And the evaluation of the Naive Bayes model uses indicators such as accuracy, precision, recall, and F1 score for comprehensive measurement. Finally, based on the performance of accuracy and F1 score on the test set, the effectiveness and robustness of the Naive Bayes model are verified.

[0066] It should be explained that in the process of obtaining the batch monitoring management set, through the hyperparameter analysis of tags and values, the joint probability of various tags (such as industry, region, user enterprise type, etc.) and values (such as sales volume, power consumption, temperature, etc.) in the data is evaluated. Hyperparameters are parameters that affect the performance of the model, including weights, biases, distribution parameters, etc. By adjusting and optimizing the hyperparameters, the relationship between data can be represented more accurately, thereby optimizing the potential allocation model and enabling better prediction and identification of the relationship between tags and values.

[0067] Optimize the potential allocation model through statistical and machine learning methods to judge the correlation between different tags and between tags and values. Specifically, techniques such as Pearson correlation coefficient, mutual information, and covariance analysis can be used to analyze the relationship between tags and between values. Based on the correlation between tags and between values, calculate the weights of each tag and value. The purpose is to judge which factors are more important for the overall operation data through quantitative analysis.

[0068] According to the determined tag weights and main body characteristics, combined with existing operation rules, allocate appropriate classification category tags to each operation information data. The classification tags include but are not limited to: industry classification, regional classification, user classification. Furthermore, it can be automatically classified and incorporated into different management, making the operation management more orderly and efficient. After obtaining the industry, regional, and user classification tags, analyze the joint distribution between the classification tags and the main body characteristics. Perform clustering analysis based on the correlation of the joint distribution. Through the clustering results and tag and numerical weights, use the Naive Bayes classification model for training. Naive Bayes is a classification method based on probability theory, learning how to predict the classification results of data according to characteristics such as industry classification, regional classification, and user classification, generating a set of classification models. Based on the training results, generate classification models for the three dimensions of industry, region, and user respectively, and conduct statistics and analysis on the operation information data.

[0069] Specifically, the industry classification model can analyze the trend of operation data according to industry characteristics; the regional classification model can analyze data such as electricity demand according to regional characteristics; the user classification model can analyze and predict individual needs according to user behavior, and generate a batch monitoring management set through the classification model.

[0070] Step S2, construct a virtual power plant electric energy trading framework based on the batch monitoring management set, analyze the difference between electricity supply and demand and the power balance relationship between virtual power plants, and output the energy dispatch result within the management scope of the virtual power plant.

[0071] In one embodiment, constructing a virtual power plant electric energy trading framework based on the batch monitoring management set, analyzing the difference between electricity supply and demand and the power balance relationship between virtual power plants, and outputting the energy dispatch result within the management scope of the virtual power plant includes:

[0072] Obtain the historical power generation of each energy within the management scope of the virtual power plant, and combine environmental factor data to predict the renewable energy generation. Construct a virtual power plant electric energy trading framework based on the renewable energy generation and electricity price;

[0073] Batch predict the load fluctuation trend of the virtual power plant management based on the batch monitoring management set, and analyze the difference between electricity supply and demand during the virtual power plant management process according to the renewable energy generation and the load fluctuation trend;

[0074] Based on the virtual power plant electric energy trading framework, evaluate the difference between electricity supply and demand and the power balance relationship between virtual power plants, and determine the dispatch mechanism of virtual power plant electric energy trading according to the power balance relationship to analyze the difference.

[0075] Specifically, in the process of obtaining the historical power generation of each energy within the management scope of the virtual power plant, combining the environmental factor data to predict the renewable energy generation, and constructing the virtual power plant electricity energy trading framework based on the renewable energy generation and electricity price, the historical power generation of each energy within the management scope of the virtual power plant can be extracted based on the grid monitoring equipment, and the environmental factor data within the management scope of the virtual power plant can be obtained according to the demand during the prediction period; define the prediction consecutive days and prediction consecutive weeks based on the historical power generation, analyze the number of power generation change periods and power generation data within the prediction consecutive days, calculate the daily power generation change factor and daily average change factor; analyze the number of power generation change periods and power generation data within the prediction consecutive weeks, calculate the weekly power generation change factor and weekly average change factor, and combine the calculation results with the environmental factor data to predict the renewable energy generation; design the virtual power plant electricity energy trading rules based on the prediction results, construct the virtual power plant electricity energy trading framework, and analyze the energy cost within the management scope of the virtual power plant under the condition of predicted renewable energy generation.

[0076] Specifically, combining the calculation results with the environmental factor data to predict the renewable energy generation includes: judging the simulated power generation during the prediction period according to the daily average change factor and weekly average change factor, and determining the influencing factors affecting the energy power generation result based on the environmental factor data; respectively considering the ratios of the daily average change factor and weekly average change factor in the simulated power generation based on the influencing factors, and analyzing the sum of squared errors of the simulated power generation according to the ratio results; determining the comprehensive weight factor by minimizing the sum of squared errors, determining the selection results of the prediction consecutive days and prediction consecutive weeks based on the comprehensive weight factor, and obtaining the predicted renewable energy generation according to the selection results; obtaining the day-ahead electricity price data, formulating the quotation rules of the virtual power plant in combination with the predicted renewable energy generation, and selecting the trading market according to the quotation results to provide a basis for the construction of the virtual power plant electricity energy trading framework.

[0077] Among them, the calculation formula for the predicted renewable energy generation of energy is:

[0078] ;

[0079] In the formula, E a,b represents the predicted renewable energy generation of energy in the b time period during the a - th acquisition process, ω represents the comprehensive weight factor, R represents the total number of prediction consecutive days, G r,b represents the power generation in the b time period of the r - th prediction consecutive day, represents the daily average power generation of the r - th prediction consecutive day, D represents the total number of prediction consecutive weeks, D d,b represents the power generation in the b time period of the d - th prediction consecutive week, represents the weekly average power generation of the d - th prediction consecutive week, represents the average power generation within the management scope of the virtual power plant.

[0080] It should be noted that during the process of constructing the virtual power plant electricity energy trading framework, historical power generation data is extracted, which can be specifically extracted from grid monitoring devices, usually at the hourly or 15-minute level, to provide high-quality training samples for subsequent prediction models. At the same time, environmental factor data including light intensity, wind speed, temperature, humidity, and weather conditions is obtained. For the future predicted day or predicted week (such as the next week or the next three days), the acquisition frequency can be adaptively modified according to actual needs, such as every 5 minutes or one hour. Environmental factors are the key driving variables for the output of renewable energy (especially wind energy and solar energy). Representative continuous time periods are selected based on historical power generation data. It is assumed that the continuous prediction days are 3 to 5 days, and the continuous prediction weeks are 1 to 2 weeks. A reference basis for future prediction is established by selecting a reasonable historical time window.

[0081] The number of time slices with obvious power changes within a day or a week is analyzed. The ratio or difference of power changes in each time period is analyzed to judge the average value of the power change factor on a certain day / week, reflecting the fluctuation trend of energy output in the time dimension, providing support for simulation generation and regression modeling. Since light has the greatest impact on the output of photovoltaic systems and wind speed has a significant impact on wind power systems, it is further clarified which environmental variables dominate the power generation of each energy source.

[0082] Through a regression model or a weighting function, consider the daily average change factor, the weekly average change factor, and the weight of the environmental impact factor to simulate the power generation value at each future moment. At the same time, calculate the sum of squared errors of the simulated power generation according to the simulated power generation to measure the difference between the simulated power generation and the true value, and minimize the sum of squared errors to determine the comprehensive weight factor to obtain the optimal comprehensive weight allocation to make the prediction more accurate. According to the weight factor with the minimum error, it is deduced which "continuous days" and "continuous weeks" of samples are more suitable for the current prediction target, and then the selection of training samples can be optimized to improve the generalization ability of the model.

[0083] Furthermore, based on the predicted renewable energy volume and the day-ahead electricity price (which can be obtained from the market interface or historical data), the rule-setting idea can be based on supply being greater than demand → reporting a low price to seize the market, and demand being greater than supply → reporting a slightly higher price to prioritize ensuring quantity. The purpose is to formulate a reasonable bidding strategy to maximize revenue or grid acceptance. During the process of selecting the trading market, the day-ahead market and the real-time market can be selected, providing decision-making support for the virtual power plant to participate in market trading.

[0084] Specifically, in the process of predicting the load fluctuation trend of virtual power plant management based on batch monitoring management set and analyzing the power supply-demand difference in the virtual power plant management process according to the renewable energy generation and the load fluctuation trend, the load simulation targets and load evaluation indicators corresponding to industry classification, regional classification, and user classification can be determined respectively based on the batch monitoring management set, and the distribution and parameters of random sampling can be set for random sampling; according to the random sampling results and statistical simulation techniques, a difference matrix is constructed to generate random operation information samples, and the attribute theory based on information theory is used to reduce the random operation information samples to construct a trend prediction model; based on the trend prediction model, the load fluctuation trends corresponding to industry classification, regional classification, and user classification are predicted respectively, and the electricity demand within the scope of virtual power grid management is analyzed according to the load fluctuation trend; the electricity demand is compared with the predicted result of the renewable energy generation, and the energy profit and loss state in the virtual power plant management process is determined according to the comparison result, and the power supply-demand difference is analyzed based on the energy profit and loss state.

[0085] Among them, in the process of constructing a difference matrix according to the random sampling results and statistical simulation techniques to generate random operation information samples, and using the attribute theory based on information theory to reduce the random operation information samples to construct a trend prediction model, the reliability of the random sampling results can be simulated and judged based on the load evaluation indicators, and the random sampling results can be optimized and adjusted according to the reliability results until the reliability meets the load simulation target to obtain random operation information samples; the random operation information samples are mapped to the attribute space to construct a difference matrix, and the data in the difference matrix are traversed and searched until an empty set is encountered and the search stops, and a reduction triangular matrix is obtained according to the search results; the original sub-attribute assignments of the random operation information samples in the reduction triangular matrix are analyzed and verified with the original sub-attribute assignments corresponding to the difference matrix, and the reduction accuracy of the reduction triangular matrix is judged according to the verification results; the random operation information samples corresponding to the reduction triangular matrix that meet the reduction requirements are used as the data set, and the data set is used as the input and the load fluctuation is used as the output to construct a trend prediction model in combination with the Monte Carlo algorithm.

[0086] It should be explained that according to industry classification, regional classification, and user classification, the corresponding load simulation targets are determined respectively, and the corresponding load evaluation indicators are set, such as maximum load, minimum load, load fluctuation range, etc. The purpose is to set different load simulation targets for different types of users (industries, regions, user types), and the load evaluation indicators are used to quantify the characteristics of load fluctuation and are the basis for subsequent prediction analysis and optimization.

[0087] Set the random sampling distribution of the load data, select a probability distribution function (including but not limited to normal distribution, Poisson distribution, etc.) to simulate the load fluctuation, and determine the parameters of random sampling, such as sample size, sampling frequency, sampling range, etc. The purpose is to use statistical simulation technology to construct an efficient sample data set for load fluctuation prediction, and obtain load data samples through random sampling to simulate the load fluctuation during the operation process.

[0088] Based on the random sampling results, use a discrimination matrix to generate random operation information samples. The discrimination matrix is used to measure the differences between different operation information samples, and evaluate the diversity of load fluctuations by comparing the differences between samples. The discrimination matrix helps to understand the similarity between different load data samples and provides a basis for further trend prediction. At the same time, the random operation information samples reflect the sample set of the simulated load fluctuation trend.

[0089] Use the attribute theory reduction method in information theory to simplify the random operation information samples, remove redundant or unimportant sample features. Through the simplified samples, attribute reduction can improve the accuracy and efficiency of the model, avoid overfitting, and the simplified samples will more effectively capture the trend of load fluctuations. It should be explained that in the process of reducing the random operation information samples based on the attribute theory of information theory, the cross-integration of rough set theory and feature selection methods based on information theory is mainly used to simplify the random operation information samples.

[0090] Judge the reliability of the random sampling results according to the load evaluation index, and evaluate whether the samples meet the expected load simulation goals. The reliability judgment ensures the representativeness of the samples and the reliability of the prediction, reduces the impact of low-quality data on the model, maps the random operation information samples to the attribute space, constructs a discrimination matrix, and conducts a traversal search. Stop the search when an empty set is encountered to obtain a reduced triangular matrix, which is an optimized data structure containing the minimum necessary information. The discrimination matrix helps to understand the differences between each data sample, and the reduced triangular matrix further simplifies and extracts the most valuable prediction information. The reduced triangular matrix can be used as the core data set for subsequent analysis, reducing unnecessary calculations and data processing.

[0091] Verify the samples in the reduced triangular matrix to ensure that the reduced data set can effectively reflect the load fluctuation trend and ensure that key features are not lost during the data reduction process. Use the optimized random operation information samples as the input and the load fluctuation as the output, and combine with the Monte Carlo algorithm for simulation and prediction. The Monte Carlo algorithm simulates the load fluctuation trend under different scenarios through multiple random samplings and calculations. Based on the output of the trend prediction model (i.e., the predicted load fluctuation trend), combine with the prediction results of the renewable energy generation to analyze the difference between power supply and demand. By comparing the load demand with the renewable energy supply, evaluate the operating status of the virtual power plant and help the dispatching system make flexible adjustments.

[0092] In step S3, the air-conditioning load data is used as a demand response resource to optimize the energy dispatch result, generate an air-conditioning control issuing instruction, and record and decompose the air-conditioning control issuing instruction to output the control state of the air conditioner.

[0093] In one embodiment, in the process of using the air-conditioning load data as a demand response resource to optimize the energy dispatch result, generate an air-conditioning control issuing instruction, and record and decompose the air-conditioning control issuing instruction to output the control state of the air conditioner, the power load data of various air conditioners within the management scope of the virtual power plant can be obtained as a demand response resource, and the demand response amount of the air-conditioning power load data during the prediction period can be judged. Based on the demand response amount, the energy dispatch result is adjusted; according to the adjusted energy dispatch result, the control strategies of various air conditioners are determined, the air-conditioning control issuing instruction is determined, and the air-conditioning control issuing instruction is sent to the device gateway layer through the message transmission protocol; the device gateway layer receives and analyzes the air-conditioning control issuing instruction, obtains the power information required for each air conditioner to adjust the control state of the air conditioner, and evaluates the effect of the execution state of the air conditioners participating in the control.

[0094] It should be explained that the complete closed-loop control process of using the air-conditioning load as a demand response resource to optimize the energy dispatch result of the virtual power plant. The core purpose is to achieve flexible response on the power demand side through the adjustability of the air conditioner and incorporate this response ability into the energy dispatch system, so as to achieve load peak shaving and valley filling, supply-demand balance and operation optimization. First, it collects the historical power load data and real-time power curves of various air conditioners (split air conditioners, VRV multi-connected units, water-cooled units, magnetic levitation chillers, intelligent variable-frequency air conditioners) within the management scope of the virtual power plant, providing basic data for subsequent prediction and regulation, identifying the fluctuation characteristics of the air-conditioning load, which is an important input for the dispatching resources. Judge the adjustable load amount of the air conditioner, that is, the demand response amount, according to the acceptable comfort change of the user during the prediction period in combination with the load data, and feedback the demand response amount of the air conditioner to the virtual power plant dispatching system to dynamically adjust the energy supply-demand matching, thereby improving the flexibility and economy of the power system dispatching. Cut the air-conditioning load during peak hours to avoid system overload; improve the utilization efficiency during low valley hours. According to the dispatching optimization result, generate personalized control strategies for each type of air conditioner, including control types (temperature setpoint change, on-off control, wind speed adjustment, etc., control duration, start time, recovery strategy, etc.), and materialize the demand response result into executable control instructions.

[0095] Convert the control strategy into standardized instructions (such as Modbus, BACnet, MQTT protocol formats). Each instruction contains the target device, control parameters, execution time, etc. The purpose is to build a standard control instruction set to ensure that the regulation commands can be received and executed by the actual system. Send the control instructions to the device gateway layer through a message transmission protocol (such as MQTT, Modbus, etc.) to achieve the command transmission path from the virtual power plant control center to the device layer.

[0096] The gateway analyzes the received control instructions, translates them into control actions that can be recognized by the device layer, obtains the control content (such as adjusting the temperature set point to 25°C) and executes it, thereby ensuring the accurate implementation of the control strategy, being compatible with different types of air-conditioning equipment, and at the same time recording the execution status of air-conditioning control in real time, including: whether it is successfully executed, actual power change, providing a data basis for subsequent control effect evaluation.

[0097] During the process of evaluating the execution status of the air conditioners participating in the control, the power regulation error and response time can be used as evaluation indicators to construct a control closed-loop mechanism. The power regulation error is used to measure the deviation between the actual air-conditioning power output and the desired control value, and the response time is used to evaluate the speed and timeliness of instruction response. Assume that during the actual measurement process, the air-conditioning status data is collected every 10 seconds. The results show that the average control response time is 14.6 seconds, the average power regulation error is ±0.23 kW, and the control success rate reaches 97.8%. This indicates that the air-conditioning execution end can respond to control signals stably and efficiently, thus ensuring the effectiveness and closed-loop performance of the overall load regulation strategy.

[0098] It should be explained that when obtaining historical power generation and air-conditioning load data, it is necessary to preprocess the historical power generation and air-conditioning load data to discover missing, noisy, outliers or jump points in the data, etc., to avoid prediction deviations.

[0099] Please refer to Figure 2 , the present invention also provides a virtual power plant energy management system based on air-conditioning load. The management system includes: A monitoring and management set generation module 1, which is used to obtain the electricity consumption data within the scope of virtual power plant management, integrate the operation information data, and make a unified classification of the operation information data according to the operation rules and potential allocation models to generate a batch monitoring and management set; An energy dispatch result output module 2, which is used to construct a virtual power plant electricity energy trading framework based on the batch monitoring and management set, analyze the difference between electricity supply and demand and the power balance relationship between virtual power plants, and output the energy dispatch result within the scope of virtual power plant management; The air conditioner control status output module 3 is used to take the air conditioner load data as a demand response resource, optimize the energy dispatch result to generate an air conditioner control instruction for distribution, and record and decompose the air conditioner control instruction for distribution, and output the control status of the air conditioner.

[0100] In summary, by means of the above technical solutions of the present invention, the present invention takes the air conditioner load data as a demand response resource, and the virtual power plant can flexibly adjust the air conditioner load according to the fluctuation of power demand, so as to balance the power supply and demand of the virtual power plant, greatly improving the adaptability of the power system during load fluctuations. At the same time, by using the electricity energy trading framework to analyze the supply-demand difference, the load change of the power system can be predicted in advance. Through the optimized dispatch of the air conditioner load, the demand for the air conditioner load can be reduced during the peak power demand period, thus effectively reducing the system load pressure and avoiding excessive energy consumption. The present invention predicts the short-term and medium-term renewable energy output by combining historical power generation with environmental factors such as weather, temperature, and season, which can cope with the strong volatility of renewable energy in advance, avoid the abandonment of electricity due to over-generation or passive emergency due to insufficient power generation. At the same time, when the electricity supply-demand difference is evaluated, the virtual power plant can adopt different dispatch mechanisms according to the balance relationship to realize the automation and intelligence of energy dispatch and improve the economic benefits of the virtual power plant.

[0101] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with this embodiment can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0102] Although the specific implementation manners of the present invention have been described above in conjunction with the accompanying drawings, it is not a limitation to the protection scope of the present invention. Those skilled in the art should understand that based on the technical solutions of the present invention, various modifications or deformations that can be made by those skilled in the art without creative efforts are still within the protection scope of the present invention.

Claims

1. A virtual power plant energy management method based on air conditioning load, characterized in that, The management method includes: Obtain the electricity consumption data within the scope of virtual power plant management, integrate the operation information data, and conduct a unified classification of the operation information data according to the operation rules and potential allocation model to generate a batch monitoring management set; Based on the batch monitoring management set, construct a virtual power plant electricity energy trading framework, analyze the electricity supply and demand difference and the power balance relationship between virtual power plants, and output the energy dispatch result within the scope of virtual power plant management; Use the air-conditioning load data as a demand response resource, optimize the energy dispatch result to generate an air-conditioning control instruction for distribution, and record and decompose the air-conditioning control instruction for distribution to output the control status of the air conditioner.

2. The virtual power plant energy management method based on air-conditioning load according to claim 1, characterized in that The unified classification of the operation information data according to the operation rules and potential allocation model to generate a batch monitoring management set includes: Analyze the joint probability of the Naive Bayes model based on the hyperparameters of the tags and values involved in the operation information data, optimize the Naive Bayes model based on the analysis result, and judge the correlation between tags and the correlation between values; Analyze the weights of tags and values respectively based on the correlation between tags and the correlation between values, output the main characteristics of the operation information data according to the weight result, and combine with the operation rules to determine the classification category tags of the operation information data. The classification category tags include industry classification, regional classification, and user classification; Analyze the joint distribution dependence between industry classification, regional classification, and user classification and the main characteristics, and cluster the dependence. Based on the clustering result, train the Naive Bayes model to obtain a potential allocation model; Based on the training result, generate classification models with industry classification, regional classification, and user classification as the themes respectively, and use the classification models to statistically analyze the operation information data according to industry, region, and user to obtain a batch monitoring management set.

3. A virtual power plant energy management method based on air conditioning load according to claim 1, characterized in that, The construction of a virtual power plant electricity energy trading framework based on the batch monitoring management set, analyzing the electricity supply and demand difference and the power balance relationship between virtual power plants, and outputting the energy dispatch result within the scope of virtual power plant management includes: Obtain the historical power generation of each energy within the scope of virtual power plant management, combine with environmental factor data to predict the renewable energy generation, and construct a virtual power plant electricity energy trading framework based on the renewable energy generation and electricity price; Based on the batch monitoring management set, batch predict the load fluctuation trend of virtual power plant management, and analyze the electricity supply and demand difference in the process of virtual power plant management according to the renewable energy generation and load fluctuation trend; Based on the virtual power plant electricity energy trading framework, evaluate the electricity supply and demand difference and the power balance relationship between virtual power plants, and determine the dispatch mechanism of virtual power plant electricity energy trading according to the power balance relationship and the difference analysis.

4. A virtual power plant energy management method based on air-conditioning load according to claim 3, characterized in that The obtaining of the historical power generation of each energy within the scope of virtual power plant management, combining with environmental factor data to predict the renewable energy generation, and constructing a virtual power plant electricity energy trading framework based on the renewable energy generation and electricity price includes: Extract the historical power generation of each energy within the scope of virtual power plant management based on grid monitoring equipment, and obtain the environmental factor data within the scope of virtual power plant management according to the demand in the prediction period; Define the prediction continuous days and prediction continuous weeks according to the historical power generation, and analyze the number of power generation change periods and power generation data within the prediction continuous days, calculate the daily power generation change factor and the daily average change factor; Analyze and predict the number of power generation change periods and power generation data within consecutive weeks, calculate the weekly power generation change factor and the weekly average change factor, and combine the calculation results with environmental factor data to predict the renewable energy generation; Based on the prediction results, design the virtual power plant's electricity energy trading rules, construct the virtual power plant's electricity energy trading framework, and analyze the energy costs within the management scope of the virtual power plant under the condition of predicted renewable energy generation.

5. A virtual power plant energy management method based on air-conditioning load according to claim 4, characterized in that The combination of the calculation results with environmental factor data to predict the renewable energy generation includes: Judge the simulated power generation of the prediction period according to the daily average change factor and the weekly average change factor, and determine the influencing factors affecting the power generation result based on the environmental factor data; Based on the influencing factors, respectively consider the ratios of the daily average change factor and the weekly average change factor in the simulated power generation, and analyze the sum of squared errors of the simulated power generation according to the ratio results; Determine the comprehensive weight factor by minimizing the sum of squared errors, determine the selection results of consecutive prediction days and consecutive prediction weeks based on the comprehensive weight factor, and obtain the predicted renewable energy generation according to the selection results; Obtain the day-ahead electricity price data, formulate the virtual power plant's bidding rules in combination with the predicted renewable energy generation, and select the trading market according to the bidding results to provide a basis for the construction of the virtual power plant's electricity energy trading framework.

6. The virtual power plant energy management method based on air-conditioning load according to claim 5, wherein, The calculation formula for the predicted renewable energy generation is: ； where E a,b represents the predicted regenerated energy of the b time period during the a-th acquisition process, ω represents the comprehensive weight factor, R represents the total number of consecutive days of prediction, G r,b represents the power generation of the b time period on the r-th consecutive day of prediction, represents the daily average power generation on the r-th consecutive day of prediction, D represents the total number of consecutive weeks of prediction, D d,b represents the power generation of the b time period in the d-th consecutive week of prediction, represents the weekly average power generation in the d-th consecutive week of prediction, represents the average power generation within the scope of virtual power plant management.

7. A virtual power plant energy management method based on air-conditioning load according to claim 6, characterized in that The batch monitoring management set is used to batch predict the load fluctuation trend managed by the virtual power plant, and the electricity supply and demand difference in the virtual power plant management process is analyzed based on the renewable energy generation and the load fluctuation trend, including: Based on the batch monitoring management set, respectively determine the load simulation targets and load evaluation indicators corresponding to industry classification, regional classification, and user classification, and set the distribution and parameters of random sampling for random sampling; Construct a difference matrix based on the random sampling results and statistical simulation techniques to generate a random operation information sample, and use the attribute theory based on information theory to reduce the random operation information sample to construct a trend prediction model; Based on the trend prediction model, respectively predict the load fluctuation trends corresponding to industry classification, regional classification, and user classification, and analyze the electricity demand within the management scope of the virtual power grid according to the load fluctuation trends; Compare the difference between the electricity demand and the predicted result of the renewable energy generation, determine the energy profit and loss status in the virtual power plant management process according to the comparison result, and analyze the electricity supply and demand difference based on the energy profit and loss status.

8. A virtual power plant energy management method based on air-conditioning load according to claim 7, characterized in that, The construction of a difference matrix based on the random sampling results and statistical simulation techniques to generate a random operation information sample, and the use of the attribute theory based on information theory to reduce the random operation information sample to construct a trend prediction model includes: Based on the load evaluation index, simulate and judge the reliability of the random sampling results, optimize and adjust the random sampling results according to the reliability results until the reliability meets the load simulation target to obtain a random operation information sample; Map the random operation information sample to the attribute space to construct a difference matrix, traverse and search the data in the difference matrix until an empty set is encountered and then stop the search, and obtain a reduced triangular matrix according to the search results; Analyze the original attribute assignment of the random operation information samples in the reduction triangular matrix, verify it with the corresponding original attribute assignment in the difference matrix, and judge the reduction accuracy of the reduction triangular matrix according to the verification result; Use the random operation information samples corresponding to the reduction triangular matrix that meet the reduction requirements as the data set, and use the data set as the input and the load fluctuation as the output to construct a trend prediction model in combination with the Monte Carlo algorithm.

9. The virtual power plant energy management method based on air-conditioning load according to claim 1, wherein The steps of using the air conditioner load data as a demand response resource, optimizing the energy dispatching result to generate an air conditioner control instruction for distribution, and recording and decomposing the air conditioner control instruction for distribution to output the control state of the air conditioner include: Obtain the power load data of various air conditioners within the management scope of the virtual power plant as demand response resources, judge the demand response amount of the air conditioner power load data during the prediction period, and adjust the energy dispatching result based on the demand response amount; Determine the control strategies of various air conditioners according to the adjusted energy dispatching result, determine the air conditioner control instruction for distribution, and send the air conditioner control instruction for distribution to the device gateway layer through the message transmission protocol; The device gateway layer receives and analyzes the air conditioner control instruction for distribution, obtains the power information required for each air conditioner to adjust the control state of the air conditioner, and evaluates the effect of the execution state of the air conditioners participating in the control.

10. A virtual power plant energy management system based on air-conditioning load, which is used to implement the virtual power plant energy management method based on air-conditioning load described in any one of claims 1-9, and is characterized in that, The management system includes: A monitoring and management set generation module, which is used to obtain the power consumption data within the management scope of the virtual power plant, integrate the operation information data, and uniformly divide the operation information data according to the operation rules and the potential allocation model to generate a batch monitoring and management set; An energy dispatching result output module, which is used to construct a virtual power plant electricity energy trading framework based on the batch monitoring and management set, analyze the electricity supply and demand difference and the power balance relationship between virtual power plants, and output the energy dispatching result within the management scope of the virtual power plant; An air conditioner control state output module, which is used to use the air conditioner load data as a demand response resource, optimize the energy dispatching result to generate an air conditioner control instruction for distribution, and record and decompose the air conditioner control instruction for distribution to output the control state of the air conditioner.

Images