Power demand side management method and system based on machine learning

Abstract

The invention is suitable for the field of power demand management, and provides a power demand side management method and system based on machine learning, and the system comprises a power grid data collection and stability analysis module, a virtual power plant simulation demand prediction module, a charging and discharging strategy optimization module, and a strategy feedback evaluation module. According to the scheme, the optimized charging and discharging strategy is generated based on the prediction result of the future power supply and demand condition in combination with the power market price and the charging and discharging efficiency of the energy storage equipment. By constructing a charging and discharging optimization model, the economical efficiency and high efficiency of the strategy are ensured by taking minimization of the charging and discharging cost and maximization of the system benefit as targets. When a charging and discharging strategy is formulated, multiple factors such as a power demand prediction result, a power market price, charging and discharging efficiency and state of energy storage equipment and the like are comprehensively considered, and scientificity and feasibility of the strategy are ensured.

Description

Technical Field

[0001] The present invention belongs to the field of power demand management, and in particular relates to a power demand side management method and system based on machine learning. Background Art

[0002] Demand Side Management (DSM) refers to guiding and controlling the electricity consumption behavior of power users through various technologies and management means to achieve optimized operation of the power system and efficient use of resources. The field of power demand management optimizes the operation of the power system, improves energy efficiency, and achieves the dual goals of energy conservation, emission reduction and economic benefits through the comprehensive use of technology, market and policy means.

[0003] Traditional power system management cannot accurately predict future changes in power supply and demand, resulting in an imbalance between supply and demand, which in turn causes stability problems and economic losses in the power system. In addition, power market prices fluctuate greatly, and traditional power management strategies are difficult to adjust dynamically, resulting in the inability to minimize costs and maximize benefits during price fluctuations. The charging and discharging strategies of energy storage equipment also lack optimization, resulting in low utilization efficiency of energy storage equipment and the inability to fully tap the potential of energy storage equipment. Summary of the invention

[0004] The purpose of the present invention is to provide a power demand side management method based on machine learning, aiming to solve the technical problems existing in the prior art identified in the background technology.

[0005] The present invention is implemented as follows: a method for power demand side management based on machine learning, the method comprising:

[0006] Collect real-time data of the power grid in various time periods, add time tags to each data, and analyze the stability of the power grid by combining the values ​​of each time tag of each type of data;

[0007] Obtain the distributed energy resource information in the current region, and establish a virtual power plant based on it. Build a supply and demand forecasting model in the virtual power plant, substitute the analysis results of the power grid stability and the real-time data of the power grid collected in the current time period into the demand forecasting model, and simulate the changes in power conditions in the future time period;

[0008] Based on the simulation results in the virtual power plant, combined with the current electricity market price and the charging and discharging efficiency of the energy storage equipment, an optimized charging and discharging strategy is generated;

[0009] The real-time data of the power grid and the market satisfaction evaluation are obtained again to evaluate the charging and discharging strategy.

[0010] As a further solution of the present invention, the analysis of power grid stability by combining the values ​​of each time tag of each type of data specifically includes:

[0011] Set the collection frequency for grid sensor equipment and smart meters to collect all data in grid operation in real time;

[0012] Add a corresponding time tag to each piece of collected data to record the collection time of the data;

[0013] Integrate all the data with time tags and analyze the stability of the power grid based on the data in each time period. Specifically:

[0014]

[0015] Among them, X t is the power data point at time t in the event sequence, which can be any quantity that changes over time, such as the demand, load, or electricity price of the power grid; c is a constant term, which represents the average level of power supply; X t-i represents the power data at the i-th time point; Indicates power data point X t-i For the current value X t The influence of; ∈tj represents the prediction error at the jth time point in the past; θ j Represents the past error term to the current value X t The impact of

[0016] X t Analyze to determine whether there is a sudden increase or decrease, as well as the magnitude of the increase or decrease, and assess the stability of the power grid.

[0017] As a further solution of the present invention, the establishment of a virtual power plant, the construction of a supply and demand forecasting model in the virtual power plant, and the simulation of changes in power conditions in future time periods specifically include:

[0018] Build a conceptual model of a virtual power plant, including the integration and management of distributed energy resources, to simulate actual power system operation;

[0019] Construct a supply and demand forecasting model, and based on the stability analysis results, bring the collected real-time data of the power grid with the current time tag into the supply and demand forecasting model to predict and analyze the changes in the future power supply and demand situation. Specifically:

[0020]

[0021] Among them, y i is the target variable, which represents the actual electricity demand at time point i; is the predicted value, which represents the predicted value of the model for the electricity demand at time point i, and is the result of the model's prediction based on the input features; l is the loss function, which is used to measure the difference between the model's predicted value and the true value.

[0022] As a further solution of the present invention, the above-mentioned generation of an optimized charging and discharging strategy by combining the current electricity market price and the charging and discharging efficiency of the energy storage device specifically includes:

[0023] Based on the forecast results of future changes in electricity supply and demand, identify the peak and trough periods of future electricity demand and determine the imbalance points of electricity supply and demand;

[0024] Obtain current electricity market price information and generate future electricity price trends based on forecast results, while monitoring the current energy storage device charging status, capacity, and efficiency parameters;

[0025] With minimizing charging and discharging costs and maximizing system benefits as the objective function, a charging and discharging optimization model is constructed. The constraints include the charging and discharging efficiency of the energy storage equipment, the power demand forecast results and the market price. A specific charging and discharging strategy is formulated and it is determined when to perform charging or discharging operations.

[0026] As a further solution of the present invention, the real-time data of the power grid and the market satisfaction evaluation are obtained to evaluate the charging and discharging strategy, which specifically includes:

[0027] Collect real-time data and grid operation status after the implementation of charging and discharging strategies, and collect market satisfaction scores after the implementation of charging and discharging strategies;

[0028] Based on the collected real-time data and market satisfaction evaluation, the implementation effect of the charging and discharging strategy is analyzed, including the balance of power load, fluctuations in electricity prices and the utilization efficiency of energy storage equipment.

[0029] Another object of the present invention is to provide a power demand side management system based on machine learning, the system comprising:

[0030] The power grid data collection and stability analysis module is used to collect real-time data of the power grid in various time periods, including load, power generation, frequency and voltage, and add a time tag to each data. It also combines the values ​​of each time tag of each type of data to analyze the stability of the power grid;

[0031] The virtual power plant simulation demand forecasting module is used to obtain the distributed energy resource information in the current region and establish a virtual power plant based on it. A supply and demand forecasting model is built in the virtual power plant. The analysis results of the power grid stability and the real-time data of the power grid collected in the current time period are substituted into the demand forecasting model to simulate the changes in the power situation in the future time period, including power demand forecast and power generation forecast;

[0032] The charging and discharging strategy optimization module is used to generate an optimized charging and discharging strategy based on the simulation results in the virtual power plant, combined with the current electricity market price and the charging and discharging efficiency of the energy storage equipment;

[0033] The strategy feedback evaluation module is used to obtain the real-time data of the power grid and the market satisfaction evaluation again to evaluate the charging and discharging strategy.

[0034] As a further solution of the present invention, the power grid data acquisition and stability analysis module includes:

[0035] The data collection tagging unit is used to set the collection frequency for the power grid sensor equipment and smart meters, collect all the data in the power grid operation in real time, and add a corresponding time tag to each piece of collected data to record the collection time of the data;

[0036] The power grid stability assessment unit is used to integrate all the data with time tags and analyze the stability of the power grid based on the data in each time period. Specifically:

[0037]

[0038] Among them, X t is the power data point at time t in the event sequence, which can be any quantity that changes over time, such as the demand, load, or electricity price of the power grid; c is a constant term, which represents the average level of power supply; X t-i represents the power data at the i-th time point; Indicates power data point X t-i For the current value X t The influence of; ∈tj represents the prediction error at the jth time point in the past; θ j Represents the past error term to the current value X t The impact of

[0039] X t Analyze to determine whether there is a sudden increase or decrease, as well as the magnitude of the increase or decrease, and assess the stability of the power grid.

[0040] As a further solution of the present invention, the virtual power plant simulation demand forecasting module includes:

[0041] A virtual power plant construction unit, which is used to construct a conceptual model of a virtual power plant, including the integration and management of distributed energy resources, and to simulate the actual power system operation;

[0042] The supply and demand forecasting unit is used to build a supply and demand forecasting model. According to the stability analysis results, the collected real-time data of the power grid with the current time tag is brought into the supply and demand forecasting model to predict and analyze the changes in the future power supply and demand situation. Specifically:

[0043]

[0044] Among them, y i is the target variable, which represents the actual electricity demand at time point i; is the predicted value, which represents the predicted value of the model for the electricity demand at time point i, and is the result of the model's prediction based on the input features; l is the loss function, which is used to measure the difference between the model's predicted value and the true value.

[0045] As a further solution of the present invention, the charging and discharging strategy optimization module includes:

[0046] A supply and demand imbalance identification unit is used to identify the peak and valley periods of future electricity demand and determine the imbalance point of electricity supply and demand based on the prediction results of future changes in electricity supply and demand conditions;

[0047] The price and energy storage status monitoring unit is used to obtain the current electricity market price information and generate the future total electricity price change trend based on the forecast results, while monitoring the current energy storage equipment charging status, capacity and efficiency parameters;

[0048] The charging and discharging optimization unit is used to build a charging and discharging optimization model with the objective function of minimizing the charging and discharging cost and maximizing the system benefit. The constraints include the charging and discharging efficiency of the energy storage equipment, the power demand forecast results and the market price. It formulates a specific charging and discharging strategy and determines when to perform charging or discharging operations.

[0049] As a further solution of the present invention, the strategy feedback evaluation module includes:

[0050] The secondary data collection unit is used to collect real-time data and grid operation status after the implementation of the charging and discharging strategy, and collect the market satisfaction rating after the implementation of the charging and discharging strategy;

[0051] The strategy effect analysis unit is used to analyze the implementation effect of the charging and discharging strategy based on the collected real-time data and market satisfaction evaluation, including the balance of power load, fluctuations in electricity prices and the utilization efficiency of energy storage equipment.

[0052] The beneficial effects of the present invention are:

[0053] This solution generates an optimized charging and discharging strategy based on the forecast results of future power supply and demand conditions, combined with power market prices and the charging and discharging efficiency of energy storage equipment. By constructing a charging and discharging optimization model, the strategy is economical and efficient with the goal of minimizing charging and discharging costs and maximizing system benefits. When formulating the charging and discharging strategy, multiple factors such as power demand forecast results, power market prices, and the charging and discharging efficiency and status of energy storage equipment are fully considered to ensure the scientificity and feasibility of the strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flowchart of a power demand side management method based on machine learning provided by an embodiment of the present invention;

[0055] Figure 2 A flow chart for analyzing power grid stability by combining the values ​​of each time tag of each type of data provided by an embodiment of the present invention;

[0056] Figure 3 A flowchart of establishing a virtual power plant, building a supply and demand forecasting model in the virtual power plant, and simulating changes in power conditions in future time periods provided in an embodiment of the present invention;

[0057] Figure 4 A flowchart for generating an optimized charging and discharging strategy by combining the current electricity market price and the charging and discharging efficiency of the energy storage device provided in an embodiment of the present invention;

[0058] Figure 5 A flowchart for obtaining real-time data of a power grid and market satisfaction evaluation and evaluating a charging and discharging strategy provided by an embodiment of the present invention;

[0059] Figure 6 A structural block diagram of a power demand side management system based on machine learning provided by an embodiment of the present invention;

[0060] Figure 7 A structural block diagram of a power grid data acquisition and stability analysis module provided in an embodiment of the present invention;

[0061] Figure 8 A structural block diagram of a virtual power plant simulation demand forecasting module provided in an embodiment of the present invention;

[0062] Fig. 9 A structural block diagram of a charge and discharge strategy optimization module provided in an embodiment of the present invention;

[0063] Fig.10 This is a structural block diagram of a strategy feedback evaluation module provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0065] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0066] Figure 1 A flow chart of a power demand side management method based on machine learning provided by an embodiment of the present invention, such as Figure 1 As shown, a method for power demand side management based on machine learning, the method comprising:

[0067] S100, collecting real-time data of the power grid in various time periods, adding a time tag to each data, and analyzing the stability of the power grid by combining the values ​​of each time tag of each type of data;

[0068] In this step, you need to set the collection frequency of the sensor equipment and smart meters to ensure that the subtle changes in the power grid status can be captured in real time. Generally, the collection frequency of important nodes in the power grid is high, in seconds or minutes, while the collection frequency of smart meters is relatively low, which can be set to collect data every 15 minutes or every hour. The specific frequency is determined according to actual needs and equipment performance.

[0069] Through sensing equipment and smart meters, key parameters such as voltage, current, power, and frequency in the operation of the power grid are collected in real time. These data can reflect the real-time operation status of the power grid. In the process of data collection, the raw data needs to be preprocessed through filtering and denoising techniques to ensure the accuracy and reliability of the data. Add an accurate timestamp to each piece of collected data and record the data collection time, so that the data has a time dimension, which is convenient for subsequent time series analysis. At the same time, ensure the time synchronization of all devices, and use a unified time source (such as GPS time or network time protocol NTP) for calibration to ensure the consistency and accuracy of the time tag.

[0070] Integrate data from different sensor devices and smart meters into a unified data platform to form a complete grid operation data set. Use machine learning and data mining techniques to analyze time-tagged data, evaluate the stability of the grid from different dimensions, and calculate the grid stability index xt. By analyzing the changing trend of xt, determine whether there are abnormal fluctuations (such as sudden changes in power demand or power supply failures), and quantify the magnitude and duration of the changes. According to the analysis results of xt, the stability of the grid is graded (such as high stability, medium stability, and unstable), and an early warning mechanism is set up. When signs of instability in the grid are detected, a warning is issued in a timely manner and necessary countermeasures are taken.

[0071] like Figure 2 As shown, the combination of the values ​​of each time tag of each type of data to analyze the stability of the power grid specifically includes:

[0072] S110, sets the collection frequency for grid sensor equipment and smart meters to collect all data in grid operation in real time;

[0073] S120, adding a corresponding time tag to each piece of collected data to record the collection time of the data;

[0074] S130 integrates all the data with time tags and analyzes the stability of the power grid based on the data in each time period. Specifically:

[0075]

[0076] Among them, X t is the power data point at time t in the event sequence, which can be any quantity that changes over time, such as the demand, load, or electricity price of the power grid; c is a constant term, which represents the average level of power supply; X t-i represents the power data at the i-th time point; Indicates power data point X t-i For the current value X t The influence of; ∈tj represents the prediction error at the jth time point in the past; θ j Represents the past error term to the current value X t The impact of

[0077] X t Analyze to determine whether there is a sudden increase or decrease, as well as the magnitude of the increase or decrease, and assess the stability of the power grid.

[0078] S200, obtaining the distributed energy resource information of the current region, and establishing a virtual power plant based on the information, building a supply and demand forecasting model in the virtual power plant, substituting the analysis results of the grid stability and the real-time grid data collected in the current time period into the demand forecasting model, and simulating the changes in the power situation in the future time period;

[0079] This step requires the construction of a conceptual model of a virtual power plant, which includes the integration and management of distributed energy resources and the simulation of the operation of the actual power system. The virtual power plant reproduces the dynamic behavior of the power system in a digital way to ensure the optimal allocation and coordinated operation of resources. The supply and demand forecasting model uses advanced machine learning algorithms, such as regression analysis, time series analysis, neural networks, etc., to accurately forecast future power demand based on historical data and real-time data. The model sets a target variable to represent the actual power demand at a specific time point, and the predicted value is the model's prediction result of the power demand at the same time point. The loss function is used to measure the difference between the model's predicted value and the true value, and the model parameters are optimized by minimizing the loss function. In the data input and forecast analysis stage, the analysis results of the grid stability and the real-time data with time tags are input into the supply and demand forecasting model. The model performs forecast analysis based on these data to simulate the changes in power supply and demand in future time periods, including the changing trend of power demand, peak periods and valley periods, etc.

[0080] like Figure 3 As shown, the establishment of a virtual power plant, the construction of a supply and demand forecasting model in the virtual power plant, and the simulation of changes in power conditions in future time periods specifically include:

[0081] S210, construct a conceptual model of a virtual power plant, including the integration and management of distributed energy resources, to simulate actual power system operation;

[0082] S220, construct a supply and demand forecasting model, and according to the stability analysis results, bring the collected real-time data of the power grid with the current time tag into the supply and demand forecasting model to conduct forecasting analysis on the changes in the future power supply and demand situation, specifically:

[0083]

[0084] Among them, y i is the target variable, which represents the actual electricity demand at time point i; is the predicted value, which represents the predicted value of the model for the electricity demand at time point i, and is the result of the model's prediction based on the input features; l is the loss function, which is used to measure the difference between the model's predicted value and the true value.

[0085] S300, based on the simulation results in the virtual power plant, combined with the current electricity market price and the charging and discharging efficiency of the energy storage equipment, generates an optimized charging and discharging strategy;

[0086] In this step, based on the forecast results of future changes in electricity supply and demand, the peak and trough periods of electricity demand and the imbalance points of electricity supply and demand are identified. This step ensures the balance of supply and demand in the power system and provides a basis for the subsequent formulation of charging and discharging strategies.

[0087] Then, the current electricity market price information is obtained, and based on the forecast results, the future total electricity price change trend is generated. This step enables the charging and discharging strategy to adapt to the dynamic changes in the market, so as to find the best charging and discharging time in price fluctuations. At the same time, this step also monitors the current charging state, capacity and efficiency parameters of the energy storage equipment to ensure the feasibility and efficiency of the charging and discharging strategy.

[0088] In terms of objective function construction, a charging and discharging optimization model is constructed with the goal of minimizing charging and discharging costs and maximizing system benefits. The model takes into account multiple constraints, including the charging and discharging efficiency of energy storage equipment, power demand forecast results, and market prices. Through comprehensive consideration of these constraints, this step can formulate a specific charging and discharging strategy and determine when to perform charging or discharging operations to minimize costs and maximize benefits.

[0089] like Figure 4 As shown, the optimization charging and discharging strategy is generated by combining the current electricity market price and the charging and discharging efficiency of the energy storage equipment, which specifically includes:

[0090] S310, based on the prediction results of the future changes in the power supply and demand conditions, identifying the peak and valley time periods of future power demand and determining the imbalance point of power supply and demand;

[0091] S320, obtaining current electricity market price information, and generating a future total electricity price change trend based on the prediction results, while monitoring the charging state, capacity, and efficiency parameters of the current energy storage device;

[0092] S330, with minimizing the charging and discharging costs and maximizing the system benefits as the objective function, constructs a charging and discharging optimization model. The constraints include the charging and discharging efficiency of the energy storage equipment, the power demand forecast results and the market price. Develops a specific charging and discharging strategy and determines when to perform charging or discharging operations.

[0093] S400, again obtains the real-time data of the power grid and the market satisfaction evaluation, and evaluates the charging and discharging strategy.

[0094] This step will first collect real-time data on the power grid after implementing the charging and discharging strategy. These data include power load, electricity price, charging and discharging status of energy storage equipment, etc., providing basic data for evaluating the effectiveness of the strategy.

[0095] At the same time, the market satisfaction scores after the implementation of the charging and discharging strategy will be collected. These scores reflect the subjective feelings and evaluations of market participants on the effectiveness of the strategy implementation. By combining real-time data and market satisfaction evaluation, the implementation effect of the charging and discharging strategy can be comprehensively analyzed, including the balance of power load, fluctuations in electricity prices, and the utilization efficiency of energy storage equipment.

[0096] During the analysis process, this step focuses on the balance of power load, evaluates whether the charging and discharging strategy effectively alleviates the imbalance of power supply and demand, and whether a smooth transition of power load is achieved during peak and trough periods. In addition, the fluctuation of electricity prices is analyzed to evaluate whether the charging and discharging strategy minimizes costs and maximizes benefits in the fluctuation of electricity prices. At the same time, this step also focuses on the utilization efficiency of energy storage equipment, evaluates whether the charging and discharging strategy effectively improves the charging and discharging efficiency of energy storage equipment, and whether the optimal charging and discharging operation is achieved under the capacity and state constraints of the energy storage equipment.

[0097] like Figure 5 As shown, the real-time data of the power grid and the market satisfaction evaluation are obtained, and the charging and discharging strategy is evaluated, which specifically includes:

[0098] S410, collecting real-time data and grid operation status after the implementation of the charging and discharging strategy, and collecting market satisfaction scores after the implementation of the charging and discharging strategy;

[0099] S420, analyzing the implementation effect of the charging and discharging strategy based on the collected real-time data and market satisfaction evaluation, including the balance of power load, fluctuation of electricity price and utilization efficiency of energy storage equipment.

[0100] like Figure 6 As shown, a power demand side management system based on machine learning, the system comprises:

[0101] The power grid data collection and stability analysis module 100 is used to collect real-time data of the power grid in various time periods, including load, power generation, frequency and voltage, and add a time tag to each data, and analyze the stability of the power grid by combining the values ​​of each time tag of each type of data;

[0102] The virtual power plant simulation demand forecasting module 200 is used to obtain the distributed energy resource information of the current region, and establish a virtual power plant based on it, build a supply demand forecasting model in the virtual power plant, substitute the analysis results of the grid stability and the real-time grid data collected in the current time period into the demand forecasting model, and simulate the changes in the power situation in the future time period, including power demand forecasting and power generation forecasting;

[0103] The charging and discharging strategy optimization module 300 is used to generate an optimized charging and discharging strategy based on the simulation results in the virtual power plant, combined with the current electricity market price and the charging and discharging efficiency of the energy storage device;

[0104] The strategy feedback evaluation module 400 is used to obtain the real-time data of the power grid and the market satisfaction evaluation again, and evaluate the charging and discharging strategy.

[0105] like Figure 7As shown, the power grid data acquisition and stability analysis module includes:

[0106] The data collection and marking unit 110 is used to set the collection frequency for the power grid sensor equipment and the smart meter, collect all the data in the power grid operation in real time, and add a corresponding time tag to each piece of collected data to record the collection time of the data;

[0107] The power grid stability evaluation unit 120 is used to integrate all the data with time tags and analyze the stability of the power grid according to the data of each time period. Specifically:

[0108]

[0109] Among them, X t is the power data point at time t in the event sequence, which can be any quantity that changes over time, such as the demand, load, or electricity price of the power grid; c is a constant term, which represents the average level of power supply; X t-i represents the power data at the i-th time point; Indicates power data point X t-i For the current value X t The influence of; ∈tj represents the prediction error at the jth time point in the past; θ j Represents the past error term to the current value X t The impact of

[0110] X t Analyze to determine whether there is a sudden increase or decrease, as well as the magnitude of the increase or decrease, and assess the stability of the power grid.

[0111] like Figure 8 As shown, the virtual power plant simulation demand forecasting module includes:

[0112] A virtual power plant construction unit 210, for constructing a conceptual model of a virtual power plant, including integration and management of distributed energy resources, and simulating actual power system operation;

[0113] The supply demand prediction unit 220 is used to construct a supply demand prediction model, and according to the stability analysis results, the collected real-time data of the power grid with the current time tag is brought into the supply demand prediction model to predict and analyze the changes in the future power supply and demand conditions. Specifically:

[0114]

[0115] Among them, y i is the target variable, which represents the actual electricity demand at time point i; is the predicted value, which represents the predicted value of the model for the electricity demand at time point i, and is the result of the model's prediction based on the input features; l is the loss function, which is used to measure the difference between the model's predicted value and the true value.

[0116] like Fig. 9 As shown, the charging and discharging strategy optimization module includes:

[0117] The supply and demand imbalance identification unit 310 is used to identify the peak and valley time periods of future power demand based on the prediction results of the future power supply and demand changes, and determine the imbalance point of power supply and demand;

[0118] The price and energy storage status monitoring unit 320 is used to obtain the current electricity market price information and generate the future total electricity price change trend based on the prediction results, while monitoring the charging status, capacity and efficiency parameters of the current energy storage equipment;

[0119] The charge and discharge optimization unit 330 is used to construct a charge and discharge optimization model with minimizing the charge and discharge cost and maximizing the system benefit as the objective function. The constraints include the charge and discharge efficiency of the energy storage device, the power demand forecast result and the market price, formulate a specific charge and discharge strategy, and determine when to perform the charge or discharge operation.

[0120] like Fig.10 As shown, the strategy feedback evaluation module includes:

[0121] The secondary data collection unit 410 is used to collect the real-time data and the operation status of the power grid after the charging and discharging strategy is implemented, and to collect the satisfaction score of the market after the charging and discharging strategy is implemented;

[0122] The strategy effect analysis unit 420 is used to analyze the implementation effect of the charging and discharging strategy, including the balance of power load, fluctuation of electricity price and utilization efficiency of energy storage equipment, based on the collected real-time data and market satisfaction evaluation.

[0123] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0124] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0125] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0127] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions 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 power demand side management based on machine learning, characterized in that: The method comprises: Collect real-time data of the power grid in various time periods, add time tags to each data, and analyze the stability of the power grid by combining the values ​​of each time tag of each type of data; Obtain the distributed energy resource information in the current region and establish a virtual power plant based on it. Build a supply and demand forecasting model in the virtual power plant, substitute the analysis results of the power grid stability and the real-time data of the power grid collected in the current time period into the demand forecasting model, and simulate the changes in power conditions in the future time period; Based on the simulation results in the virtual power plant, combined with the current electricity market price and the charging and discharging efficiency of the energy storage equipment, an optimized charging and discharging strategy is generated; Once again, real-time data of the power grid and market satisfaction evaluation are obtained to evaluate the charging and discharging strategies.

2. The method according to claim 1, characterized in that The analysis of grid stability by combining the values ​​of each time tag of each type of data specifically includes: Set the collection frequency for grid sensor equipment and smart meters to collect all data in grid operation in real time; Add a corresponding time tag to each piece of collected data to record the collection time of the data; Integrate all the data with time tags and analyze the stability of the power grid based on the data in each time period. Specifically: Among them, X t is the power data point at time t in the event sequence, which can be any quantity that changes over time, such as the demand, load, or electricity price of the power grid; c is a constant term, which represents the average level of power supply; X t-i represents the power data at the i-th time point; Indicates power data point X t-i For the current value X t The influence of; ∈tj represents the prediction error at the jth time point in the past; θ j Represents the past error term to the current value X t The impact of X t Analyze to determine whether there is a sudden increase or decrease, as well as the magnitude of the increase or decrease, and assess the stability of the power grid.

3. The method according to claim 2, characterized in that The establishment of a virtual power plant, the construction of a supply and demand forecasting model in the virtual power plant, and the simulation of changes in power conditions in future time periods specifically include: Build a conceptual model of a virtual power plant, including the integration and management of distributed energy resources, to simulate actual power system operation; Construct a supply and demand forecasting model, and based on the stability analysis results, bring the collected real-time data of the power grid with the current time tag into the supply and demand forecasting model to predict and analyze the changes in the future power supply and demand situation. Specifically: Among them, y i is the target variable, which represents the actual electricity demand at time point i; is the predicted value, which represents the predicted value of the model for the electricity demand at time point i, and is the result of the model's prediction based on the input features; l is the loss function, which is used to measure the difference between the model's predicted value and the true value.

4. The method according to claim 3, characterized in that The above mentioned method combines the current electricity market price and the charging and discharging efficiency of the energy storage device to generate an optimized charging and discharging strategy, specifically including: Based on the forecast results of future changes in electricity supply and demand, identify the peak and trough periods of future electricity demand and determine the imbalance points of electricity supply and demand; Obtain current electricity market price information and generate future electricity price trends based on forecast results, while monitoring the current energy storage device charging status, capacity, and efficiency parameters; With minimizing charging and discharging costs and maximizing system benefits as the objective function, a charging and discharging optimization model is constructed. The constraints include the charging and discharging efficiency of the energy storage equipment, the power demand forecast results and the market price. A specific charging and discharging strategy is formulated and it is determined when to perform charging or discharging operations.

5. The method according to claim 4, characterized in that The real-time data of the power grid and the market satisfaction evaluation are obtained, and the charging and discharging strategy is evaluated, specifically including: Collect real-time data and grid operation status after the implementation of charging and discharging strategies, and collect market satisfaction scores after the implementation of charging and discharging strategies; Based on the collected real-time data and market satisfaction evaluation, the implementation effect of the charging and discharging strategy is analyzed, including the balance of power load, fluctuations in electricity prices and the utilization efficiency of energy storage equipment.

6. A power demand side management system based on machine learning, characterized in that: The system comprises: The power grid data collection and stability analysis module is used to collect real-time data of the power grid in various time periods, including load, power generation, frequency and voltage, and add a time tag to each data. It also combines the values ​​of each time tag of each type of data to analyze the stability of the power grid; The virtual power plant simulation demand forecasting module is used to obtain the distributed energy resource information in the current region and establish a virtual power plant based on it. A supply and demand forecasting model is built in the virtual power plant. The analysis results of the power grid stability and the real-time data of the power grid collected in the current time period are substituted into the demand forecasting model to simulate the changes in the power situation in the future time period, including power demand forecast and power generation forecast; The charging and discharging strategy optimization module is used to generate an optimized charging and discharging strategy based on the simulation results in the virtual power plant, combined with the current electricity market price and the charging and discharging efficiency of the energy storage equipment; The strategy feedback evaluation module is used to obtain the real-time data of the power grid and the market satisfaction evaluation again to evaluate the charging and discharging strategy.

7. The system according to claim 6, characterized in that The power grid data acquisition and stability analysis module includes: The data collection tagging unit is used to set the collection frequency for the power grid sensor equipment and smart meters, collect all the data in the power grid operation in real time, and add a corresponding time tag to each piece of collected data to record the collection time of the data; The power grid stability assessment unit is used to integrate all the data with time tags and analyze the stability of the power grid based on the data in each time period. Specifically: Among them, X t is the power data point at time t in the event sequence, which can be any quantity that changes over time, such as the demand, load, or electricity price of the power grid; c is a constant term, which represents the average level of power supply; X t-i represents the power data at the i-th time point; Indicates power data point X t-i For the current value X t The influence of; ∈tj represents the prediction error at the jth time point in the past; θ j Represents the past error term to the current value X t The impact of X t Analyze to determine whether there is a sudden increase or decrease, as well as the magnitude of the increase or decrease, and assess the stability of the power grid.

8. The system according to claim 7, characterized in that The virtual power plant simulation demand forecasting module includes: A virtual power plant construction unit, which is used to construct a conceptual model of a virtual power plant, including the integration and management of distributed energy resources, and to simulate the actual power system operation; The supply and demand forecasting unit is used to build a supply and demand forecasting model. According to the stability analysis results, the collected real-time data of the power grid with the current time tag is brought into the supply and demand forecasting model to predict and analyze the changes in the future power supply and demand situation. Specifically: Among them, y i is the target variable, which represents the actual electricity demand at time point i; is the predicted value, which represents the predicted value of the model for the electricity demand at time point i, and is the result of the model's prediction based on the input features; l is the loss function, which is used to measure the difference between the model's predicted value and the true value.

9. The system according to claim 8, characterized in that The charging and discharging strategy optimization module includes: A supply and demand imbalance identification unit is used to identify the peak and valley periods of future electricity demand and determine the imbalance point of electricity supply and demand based on the prediction results of future changes in electricity supply and demand conditions; The price and energy storage status monitoring unit is used to obtain the current electricity market price information and generate the future total electricity price change trend based on the forecast results, while monitoring the current energy storage equipment charging status, capacity and efficiency parameters; The charging and discharging optimization unit is used to build a charging and discharging optimization model with the objective function of minimizing the charging and discharging cost and maximizing the system benefit. The constraints include the charging and discharging efficiency of the energy storage equipment, the power demand forecast results and the market price. It formulates a specific charging and discharging strategy and determines when to perform charging or discharging operations.

10. The system according to claim 9, characterized in that The strategy feedback evaluation module includes: The secondary data collection unit is used to collect real-time data and grid operation status after the implementation of the charging and discharging strategy, and collect the market satisfaction rating after the implementation of the charging and discharging strategy; The strategy effect analysis unit is used to analyze the implementation effect of the charging and discharging strategy based on the collected real-time data and market satisfaction evaluation, including the balance of power load, fluctuations in electricity prices and the utilization efficiency of energy storage equipment.

Images