Residential electricity consumption prediction and dispatching system and method based on big data

Abstract

The present invention discloses a system and method for predicting and dispatching residential electricity consumption based on big data, and belongs to the technical field of electric power grid. The present invention solves the problems of inaccurate existing prediction methods and difficult implementation of dispatching methods. By constructing a neural network model for predicting electricity consumption and a threshold for electricity consumption, the electricity price of residential areas is graded according to time periods, and then the power supply mode of residential areas is optimized to establish a renewable backup power supply; at the same time, multiple simulation scenarios of emergencies and power failures are preset, and emergency mechanism plans are established in advance for emergencies and power failures; when the neural network model predicts that the electricity consumption of residential areas exceeds the threshold, the power supply mode is dispatched according to the predicted time point to balance the supply and demand relationship; the residential areas adopt a time-of-use electricity price policy during the off-peak period to guide residents to reduce electricity consumption; thereby reasonably allocating residential power resources and optimizing the supply and demand relationship of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power grids, and in particular to a system and method for predicting and dispatching residential electricity consumption based on big data. Background Art

[0002] The power system is one of the most important application areas of big data technology. With the rapid development of smart grids, smart meters collect a large amount of residential electricity consumption data. The electricity consumption data hides the residents' electricity consumption habits. Using data mining algorithms to mine and analyze these electricity consumption data can help the power grid to personalize the classification of residents and provide them with better services.

[0003] However, the traditional method of predicting residential electricity consumption collects and analyzes residential electricity consumption data to obtain a daily electricity consumption threshold, compares the real-time data with the threshold, and obtains the prediction result; although it can meet the prediction needs, its reference data is relatively single, and it is difficult to grasp the daily electricity consumption patterns of residents; secondly, after predicting their electricity consumption, the scheduling method is inaccurate and difficult to implement, which reduces the accuracy and practicality of residential electricity load prediction.

[0004] Therefore, the existing needs are not met, and a residential electricity consumption prediction and scheduling system and method based on big data are proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a residential electricity consumption prediction and scheduling system and method based on big data. By constructing an electricity consumption prediction neural network model and electricity consumption threshold, the electricity prices in residential areas are graded according to time periods, and then the power supply mode of the residential areas is optimized to establish a renewable backup power supply; at the same time, a variety of simulation scenarios of emergencies and power failures are preset, and emergency mechanism plans are established in advance for emergencies and power failures; when the neural network model predicts that the electricity consumption of the residential area exceeds the threshold, the power supply mode is scheduled according to the predicted time point to balance the supply and demand relationship; the residential area adopts a time-of-use electricity price policy during the off-peak period to guide residents to reduce electricity consumption, thereby solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a residential electricity consumption prediction and dispatching system based on big data, comprising:

[0007] Data acquisition module for

[0008] Access the power supply system of the residential area, obtain the historical power consumption data of the residential area according to the power supply system, collect the types of power-using equipment in the residential area and the power consumption behavior information of users, and clean the historical power consumption data;

[0009] Power consumption forecasting module for

[0010] The historical electricity consumption data is pre-processed before testing, and the pre-processed historical electricity consumption data and user electricity consumption behavior information are sequentially imported into the neural network model for training and testing to obtain the electricity consumption prediction model; the real-time electricity consumption data of residents is then imported into the neural network model for prediction to obtain the residents' electricity consumption prediction results;

[0011] Power consumption analysis module for

[0012] Analyze the user's electricity consumption behavior information and the type of electricity-using equipment in the residential area, divide the residential area's electricity consumption time period according to the user's electricity consumption behavior information, and divide the residential area's electricity consumption into grades according to the type of electricity-using equipment;

[0013] Scheduling processing module, used for

[0014] Obtain the distribution of power supply energy in residential areas in real time, and dispatch the power supply demand of residential areas accordingly according to the power consumption period and power consumption level of residential areas, so as to avoid the situation where the power supply load in residential areas is too heavy or too light.

[0015] Furthermore, the data acquisition module includes:

[0016] Data acquisition unit for

[0017] Smart meters are deployed in residential areas, and collection time nodes and durations are established for electricity consumption in residential areas. Smart meters collect electricity consumption data of each user in the residential area at scheduled times;

[0018] Cleaning conversion unit for

[0019] Abnormal data values ​​in the collected historical electricity consumption data and real-time electricity consumption data are deleted, the missing data are filled according to the average value method, and then the data is normalized to eliminate the differences between different regions and time scales. The processed electricity consumption data is transmitted to the electricity consumption prediction module for prediction based on wireless transmission technology.

[0020] Furthermore, the power consumption prediction module includes:

[0021] Pre-processing unit for

[0022] The feature extractor is used to extract effective feature information from historical power consumption data and user power consumption behavior information, and then the historical power consumption data is divided into a training set and a data set;

[0023] Model building unit for

[0024] A neural network model is constructed based on the extracted effective feature information, and the training set is imported into the neural network model for training, and the model parameters are estimated and optimized; the test set is then imported into the optimized neural network model for testing, and the prediction performance of the neural network model is evaluated;

[0025] Result judgment unit, used for

[0026] Construct a power consumption threshold, compare the prediction result with the threshold, and determine whether the power load in the residential area is normal.

[0027] Furthermore, the scheduling processing module includes:

[0028] Energy storage unit for

[0029] Introducing renewable energy as backup power, including solar panels and wind power;

[0030] Mode selection unit for

[0031] Analyze the user's electricity consumption behavior information and classify the electricity prices in residential areas according to time periods; then analyze the types of electrical equipment in residential areas and adjust the power supply mode during peak periods in residential areas based on the electricity demand in different time periods;

[0032] Emergency treatment unit for

[0033] Various simulation scenarios of emergencies and power failures are preset, and emergency mechanism plans are established in advance for emergencies and power failures.

[0034] Furthermore, it also includes:

[0035] Human-computer interaction module for

[0036] Configure a control console and display on the client to display the process and results of neural network model prediction in real time.

[0037] Furthermore, the data acquisition unit includes:

[0038] Data storage unit for

[0039] The historical electricity consumption data of residential areas and the real-time electricity consumption data of each user are stored as an electricity database.

[0040] Furthermore, the model building unit includes:

[0041] Model optimization unit for

[0042] By selecting a set of real-time electricity consumption data, the prediction performance of the neural network model is applied. After the prediction results are consistent with the actual results, the neural network model is applied to daily prediction work.

[0043] Furthermore, the power supply modes in the mode selection unit include a substation power supply mode, a reserve power supply mode, and a power supply mode combining a substation and a reserve power supply.

[0044] The method for predicting and dispatching residential electricity consumption based on big data includes the following steps:

[0045] S1. Collect historical electricity consumption data and electricity consumption behavior information of residential areas, extract electricity data features and electricity consumption behavior features after cleaning, and then divide the data into training set and test set;

[0046] S2. Build a neural network model based on the feature information, import the training set into the neural network model for training, and confirm the parameters of the neural network model; then import the test set into the neural network model for testing, and optimize the parameters of the neural network model;

[0047] S3, using smart meters to regularly collect real-time electricity consumption in residential areas, importing the real-time electricity consumption into a neural network model for prediction and outputting the prediction results;

[0048] S4. Compare the prediction result with the threshold. If the prediction result is within the threshold range, it indicates that the current power load in the residential area is normal and there is no need to adjust the power supply mode. If the prediction result exceeds the threshold range, it indicates that the current power load in the residential area is abnormal, and the corresponding power supply mode is selected according to the prediction time point.

[0049] Furthermore, after the neural network model is trained by the training set in S2, any set of real-time electricity consumption data is selected and imported into the neural network model for prediction and the prediction result is output until the prediction result is consistent with the actual result.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention constructs a power consumption prediction neural network model and a power consumption threshold, divides the electricity prices of residential areas into grades according to time periods, optimizes the power supply mode of the residential areas, and establishes a renewable backup power supply; at the same time, a variety of simulation scenarios of emergencies and power failures are preset, and emergency mechanism plans are established in advance for emergencies and power failures; when the neural network model predicts that the power consumption of the residential area exceeds the threshold, the power supply mode is dispatched according to the predicted time point to balance the supply and demand relationship; the residential area adopts a time-of-use electricity price policy during the off-peak period to guide residents to reduce electricity consumption; thereby rationally allocating residents' power resources and optimizing the supply and demand relationship of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a composition diagram of the residential electricity consumption prediction and dispatching system based on big data of the present invention;

[0053] Figure 2 This is a module diagram of the residential electricity consumption prediction and dispatching system based on big data of the present invention;

[0054] Figure 3 This is a flowchart of the steps of the method for predicting and scheduling residential electricity consumption based on big data of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] In order to solve the technical problems that the existing residential electricity load forecasting methods have relatively simple reference data and are difficult to grasp the daily electricity consumption patterns of residents; secondly, after the electricity consumption is predicted, the dispatching method is not accurate enough and difficult to implement, thus reducing the accuracy and practicality of residential electricity load forecasting, please refer to Figure 1-Figure 3 This embodiment provides the following technical solution

[0057] In this embodiment, if a city needs to predict the daily electricity consumption of a residential area, it can reasonably allocate the residents' electricity resources to ensure that the current power supply always meets the electricity demand of the residential area.

[0058] Residential electricity consumption prediction and dispatching system based on big data, including:

[0059] Data acquisition module for

[0060] The power supply system of the residential area is accessed, and the historical electricity consumption data of the residential area is obtained according to the power supply system. Then, the types of electrical equipment in the residential area and the information on the users' electricity consumption behaviors are collected, and the historical electricity consumption data is cleaned. In this embodiment, if the power supply system of the residential area can be the superior power supply department of the residential area, the historical electricity consumption data of the residential area in the past years is obtained through consultation with the power supply department. At the same time, the types of electrical equipment commonly used in the homes of users in the residential area and the peak and off-peak periods of electricity consumption in each home can be obtained through questionnaire surveys or home visits.

[0061] Data acquisition module, including:

[0062] Data acquisition unit for

[0063] Smart meters are deployed in residential areas, and collection time nodes and durations are established for electricity consumption in residential areas. Smart meters collect electricity consumption data of each user in the residential area at scheduled times. Following the above embodiment, if it is determined that all electricity meters used in the residential area are smart meters, the smart meters are used to collect statistics on daily electricity consumption data of each user. Secondly, based on working days and non-working days, the collection nodes on working days are set to twice a day, specifically 1200am and 1200pm, and the collection nodes on non-working days are set to three times a day, specifically 1000am, 1600pm and 2300pm, and each collection duration is 5 minutes. Smart meters collect electricity consumption data of each user at scheduled times according to the set time nodes and durations, and transmit the transmitted data to the next department for processing based on wireless transmission technology.

[0064] Data acquisition unit, including:

[0065] Data storage unit for

[0066] Store the historical electricity consumption data of residential areas and the real-time electricity consumption data of each user as an electricity database to provide effective data for the subsequent optimization of the neural network model;

[0067] Cleaning conversion unit for

[0068] Abnormal data values ​​in the collected historical electricity consumption data and real-time electricity consumption data are deleted, missing data are filled according to the average value method, and then the data is normalized to eliminate differences in different regions and time scales. The processed electricity consumption data is transmitted to the electricity consumption prediction module for prediction based on wireless transmission technology to ensure data quality and accuracy during the analysis process.

[0069] Power consumption forecasting module for

[0070] The historical electricity consumption data is pre-processed before testing, and the pre-processed historical electricity consumption data and user electricity consumption behavior information are sequentially imported into the neural network model for training and testing to obtain the electricity consumption prediction model; the real-time electricity consumption data of residents is then imported into the neural network model for prediction to obtain the residents' electricity consumption prediction results;

[0071] The electricity consumption prediction module includes:

[0072] Pre-processing unit for

[0073] The effective feature information in the historical electricity consumption data and the user's electricity consumption behavior information is extracted through the feature extractor, and then the historical electricity consumption data is divided into a training set and a data set; specifically, the traditional electricity consumption data prediction method is mainly based on historical data and statistical models, but with the development of big data and machine learning, the prediction method that integrates user electricity consumption behavior information data and electricity consumption data has become more accurate and reliable; after obtaining the user's electricity consumption behavior information and electricity consumption data, each data is analyzed and feature extracted, and by analyzing the relationship between the user's electricity consumption behavior information and electricity consumption data, relevant features are extracted to provide a basis for the establishment of the neural network model; among them, commonly used features include electricity consumption, electricity consumption time, user type, etc.

[0074] Model building unit for

[0075] A neural network model is constructed based on the extracted effective feature information, and the training set is imported into the neural network model for training, and the model parameters are estimated and optimized; the test set is then imported into the optimized neural network model for testing to evaluate the prediction performance of the neural network model; following the above embodiment, the types of commonly used electrical equipment in the homes of each user in the residential area and the peak and off-peak periods of electricity consumption in each home are obtained through questionnaire surveys or home visits. By analyzing the peak and off-peak periods of electricity consumption in each home, the user's electricity consumption habits, lifestyles, etc. can be understood; after the feature extraction is completed, a neural network model, i.e., a multivariate regression model is established to predict the user's electricity consumption data, and the training set and the test set are successively imported into the model for training and testing, and the relationship between the independent variable and the dependent variable is determined by regression analysis. In this method, the electricity consumption data is the dependent variable, and the user's electricity consumption behavior information is the independent variable; through regression analysis, the degree of influence of the user's electricity consumption behavior information on the electricity consumption data can be determined.

[0076] Model building unit, including:

[0077] Model optimization unit for

[0078] By selecting a set of real-time electricity consumption data to apply the prediction performance of the neural network model, and after the prediction results are consistent with the actual results, the neural network model is applied to daily prediction work; following the above embodiment, after performing regression analysis on the neural network model, in order to improve the accuracy of the model prediction, a set of electricity consumption data currently collected by the smart meter can be selected, and the real-time electricity consumption data can be imported into the neural network model for prediction and then output, and it is analyzed whether the prediction results are consistent with the actual results, thereby improving the accuracy of the neural network model prediction.

[0079] Result judgment unit, used for

[0080] Construct a power consumption threshold, compare the prediction result with the threshold, and determine whether the power load in the residential area is normal; Continuing with the above embodiment, when the training set and the test set are successively imported into the model for training and testing, multiple groups of test result data are collected, and the average value of the output results is taken out according to the average value method. This is used as the power consumption threshold to determine the accuracy of the later prediction of the neural network model.

[0081] Power consumption analysis module for

[0082] Analyze users' electricity usage behavior information and types of electricity-using equipment in residential areas, divide residential electricity usage time periods into time periods based on users' electricity usage behavior information, and divide residential electricity consumption into levels based on types of electricity-using equipment. Specifically, by analyzing users' electricity usage data and personal electricity usage behavior information, we can understand users' energy needs and preferences, provide users with suggestions for improving their electricity usage behavior, help users reasonably arrange their electricity usage time and methods, optimize energy consumption structure, and improve energy utilization efficiency. In addition, we can also provide users with real-time energy monitoring and feedback to help users understand their own energy consumption and encourage users to form the habit of saving electricity.

[0083] Scheduling processing module, used for

[0084] Obtain the distribution of power supply energy in residential areas in real time, and dispatch the power supply demand of residential areas accordingly according to the power consumption period and power consumption level of residential areas, so as to avoid the situation where the power supply load in residential areas is too heavy or too light;

[0085] Scheduling processing module, including:

[0086] Energy storage unit for

[0087] Introduce renewable energy as a backup power source. Renewable energy includes solar panels and wind power generation. Specifically, by introducing renewable energy, excess electricity can be stored and released when needed. The electricity released by the energy storage equipment can be put into the power supply system during peak hours to meet the electricity needs of residents, thereby reducing dependence on traditional power systems and reducing power supply pressure. Secondly, the introduction of renewable energy can reduce traditional electricity consumption and environmental pollution, improve energy utilization efficiency, reduce overall electricity consumption, and achieve the goal of energy conservation and emission reduction.

[0088] Mode selection unit for

[0089] The user's electricity consumption behavior information is analyzed, and the electricity prices in the residential areas are graded according to the time period; the types of electricity-using equipment in the residential areas are analyzed, and the power supply mode in the residential areas during the peak period is adjusted in combination with the electricity demand in the residential areas during different time periods; the power supply modes in the mode selection unit include the substation power supply mode, the reserve power supply mode and the power supply mode combining the substation and the reserve power supply; following the above embodiment, by analyzing the electricity consumption data collected by the smart meter, it is learned that the office workers have a peak electricity consumption period from 1900 to 2400 on weekdays, a flat peak period from 1:00 to 18:00 the next day, and a low peak period from 8:00 to 23:00 during the day on non-working days. During the peak electricity consumption period, the off-peak period is from 2400 in the evening to 700 the next day; the peak electricity consumption period for elderly users and non-office users is from 10:00 to 20:00 during the day on weekdays and non-working days, and the off-peak period is from 2100 in the evening to 900 the next day; therefore, during the peak period, the user's power supply mode can be dispatched to a power supply mode that combines substations and reserve power sources to balance the supply and demand relationship and ensure the user's electricity demand during this period; during the off-peak period, a time-of-use electricity price policy is adopted, such as 2400 in the evening as the critical point, between 2200 and 2400 on the same day, the electricity price is 0.5 cents / 1 kWh; between 2400 and 700 the next day, the electricity price is 0.3 cents / 1 kWh, thereby guiding residents to reduce electricity consumption.

[0090] Emergency treatment unit for

[0091] A variety of simulation scenarios of emergencies and power failures are preset, and emergency mechanism plans are established in advance for emergencies and power failures; following the above embodiment, for example, a scenario where the power supply system fails due to overload is simulated, and an emergency plan is established based on the scenario to access a backup power supply, use solar energy and wind power to generate electricity, and ensure the electricity demand of residential areas.

[0092] Human-computer interaction module for

[0093] Configure a control console and display on the client to display the process and results of neural network model prediction in real time.

[0094] In order to better demonstrate the operation process of the residential electricity consumption prediction and dispatching system based on big data, the present invention provides a method for the residential electricity consumption prediction and dispatching system based on big data, including the following steps:

[0095] S1. Collect historical electricity consumption data and electricity consumption behavior information of residential areas, extract electricity data features and electricity consumption behavior features after cleaning, and then divide the data into training set and test set;

[0096] S2. Build a neural network model based on the feature information, import the training set into the neural network model for training, and confirm the parameters of the neural network model; then import the test set into the neural network model for testing, and optimize the parameters of the neural network model; then select any set of real-time power consumption data and import it into the neural network model for prediction and output the prediction results until the prediction results are consistent with the actual results;

[0097] S3, using smart meters to regularly collect real-time electricity consumption in residential areas, importing the real-time electricity consumption into a neural network model for prediction and outputting the prediction results;

[0098] S4. Compare the prediction result with the threshold. If the prediction result is within the threshold range, it indicates that the current power load in the residential area is normal and there is no need to adjust the power supply mode. If the prediction result exceeds the threshold range, it indicates that the current power load in the residential area is abnormal, and the corresponding power supply mode is selected according to the prediction time point.

[0099] The beneficial effects achieved by the above content: Through the above method, the traditional power supply system and renewable energy can cooperate with each other, reasonably dispatch grid resources, optimize the supply and demand relationship of the grid, reduce the load pressure of the grid, and improve the operation efficiency and stability of the grid.

[0100] Working principle: By constructing a power consumption prediction neural network model and power consumption threshold, the electricity prices in residential areas are divided into levels according to time periods, and then the power supply mode in residential areas is optimized to establish a renewable backup power supply; at the same time, a variety of simulation scenarios for emergencies and power failures are preset, and emergency mechanism plans are established in advance for emergencies and power failures; when the neural network model predicts that the power consumption in the residential area exceeds the threshold, the power supply mode is dispatched according to the predicted time point to balance the supply and demand relationship; the residential area adopts a time-of-use electricity price policy during the off-peak period to guide residents to reduce electricity consumption; thereby rationally allocating residents' power resources and optimizing the supply and demand relationship of the power grid.

[0101] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0102] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. Residential electricity consumption forecasting and dispatching system based on big data, including: The data acquisition module is used to access the power supply system of the residential area, obtain the historical power consumption data of the residential area according to the power supply system, collect the types of power-using equipment in the residential area and the power consumption behavior information of users, and clean the historical power consumption data; The electricity consumption prediction module is used to pre-process the historical electricity consumption data before testing, and import the pre-processed historical electricity consumption data and user electricity consumption behavior information into the neural network model for training and testing to obtain the electricity consumption prediction model; then import the residents' real-time electricity consumption data into the neural network model for prediction to obtain the residents' electricity consumption prediction results; The power consumption analysis module is used to analyze the user's power consumption behavior information and the type of power consumption equipment in the residential area, divide the power consumption time periods in the residential area according to the user's power consumption behavior information, and divide the power consumption in the residential area into grades according to the type of power consumption equipment; The dispatch processing module is used to obtain the distribution of power supply energy in the residential area in real time, and to dispatch the power supply demand of the residential area accordingly according to the power consumption period and power consumption level of the residential area, so as to avoid the situation where the power supply load of the residential area is too heavy or too light; The scheduling processing module includes: The power storage unit is used to introduce renewable energy as backup power source, including solar panels and wind power generation; The mode selection unit is used to analyze the user's electricity consumption behavior information and classify the electricity prices in the residential area according to the time period; then analyze the types of electrical equipment in the residential area, and adjust the power supply mode in the peak period of the residential area in combination with the electricity demand in the residential area at different time periods; the power supply modes in the mode selection unit include: substation power supply mode, reserve power supply mode and power supply mode combining substation and reserve power supply; The emergency handling unit is used to preset multiple simulation scenarios of emergencies and power failures, and to establish emergency mechanism plans in advance for emergencies and power failures; The data acquisition module comprises: The data collection unit is used to deploy smart meters in residential areas and establish collection time nodes and durations for electricity consumption in residential areas. The smart meters collect electricity consumption data of each user in the residential area at a scheduled time. The cleaning conversion unit is used to delete abnormal data values ​​in the collected historical electricity consumption data and real-time electricity consumption data, fill in the missing data according to the average method, and then normalize the data to eliminate the differences between different regions and time scales, and transmit the processed electricity consumption data to the electricity consumption prediction module for prediction based on wireless transmission technology.

2. The residential electricity consumption prediction and dispatching system based on big data according to claim 1 is characterized by: The electricity consumption prediction module comprises: A preprocessing unit, used to extract effective feature information from historical power consumption data and user power consumption behavior information through a feature extractor, and then divide the historical power consumption data into a training set and a data set; The model building unit is used to build a neural network model based on the extracted effective feature information, import the training set into the neural network model for training, estimate and optimize the model parameters; then import the test set into the optimized neural network model for testing, and evaluate the prediction performance of the neural network model; The result judgment unit is used to construct a power consumption threshold, compare the prediction result with the threshold, and judge whether the power load in the residential area is normal.

3. The residential electricity consumption prediction and dispatching system based on big data according to claim 1 is characterized by: Also includes: The human-computer interaction module is used to configure the control console and display on the client side to display the process and results of the neural network model prediction in real time.

4. The residential electricity consumption prediction and dispatching system based on big data according to claim 1 is characterized by: The data collection unit includes: a data storage unit for storing historical power consumption data of residential areas and real-time power consumption data of each user as a power database.

5. The residential electricity consumption prediction and dispatching system based on big data according to claim 2 is characterized by: The model building unit comprises: Model optimization unit for By selecting a set of real-time electricity consumption data, the prediction performance of the neural network model is applied. After the prediction results are consistent with the actual results, the neural network model is applied to daily prediction work.

6. A method for predicting and dispatching residential electricity consumption based on big data, implemented based on the system for predicting and dispatching residential electricity consumption based on big data according to any one of claims 1 to 5, characterized in that: The following steps are included: S1. Collect historical electricity consumption data and electricity consumption behavior information of residential areas, extract electricity data features and electricity consumption behavior features after cleaning, and then divide the data into training set and test set; S2. Build a neural network model based on the feature information, import the training set into the neural network model for training, and confirm the parameters of the neural network model; then import the test set into the neural network model for testing, and optimize the parameters of the neural network model; S3, using smart meters to regularly collect real-time electricity consumption in residential areas, importing the real-time electricity consumption into a neural network model for prediction and outputting the prediction results; S4, comparing the prediction result with the threshold value. If the prediction result is within the threshold range, it indicates that the current power load of the residential area is normal and there is no need to adjust the power supply mode; If the prediction result exceeds the threshold range, it indicates that the current power load in the residential area is abnormal, and the corresponding power supply mode is selected according to the predicted time point.

7. The method for predicting and dispatching residential electricity consumption based on big data according to claim 6 is characterized by: After the neural network model is trained by the training set in S2, any set of real-time electricity consumption data is selected and imported into the neural network model for prediction and the prediction result is output until the prediction result is consistent with the actual result.

Images