Research On Intelligent Household Electricity Management System Based On Deep Learning Algorithm

With the rapid development of smart grid technology in China,intelligent household electricity consumption has become an important part of smart grid system.At present,the way to feedback users’ electricity consumption data is mainly through the smart electricity meter at the household side to measure the electricity consumption data and upload it to the cloud server.Customers can query the overall electricity consumption level of the family through the network.In this way,as the number of users increases,the efficiency of the server decreases,and users are unable to obtain detailed electricity consumption information and trends.Aiming at this problem,this paper proposes an intelligent household electricity management system for home users.The system collects the power consumption characteristics of each room of the household users through the intelligent acquisition module,and then extracts the power consumption characteristics and stores them in the local edge computing module.At the same time,the edge computing module can run the deep learning algorithm,use the household electricity consumption data of normal electricity consumption as the training sample for model training,and use the collected electricity consumption information as the input variable of the model to predict the future electricity consumption trend in different periods.Finally,the user’s historical electricity consumption data and future electricity consumption trend will be fed back in the form of mobile phone APP.According to the historical data and the future trend of electricity consumption,users can change their electricity consumption behavior to reduce electricity costs and improve the efficiency of electricity utilization.The main work of this paper is as follows:1.Make intelligent acquisition equipment through schematic design and PCB design,which is composed of acquisition circuit,metering chip,STM32 control chip and ZigBee communication module.It can accurately collect various electricity data in the family room,and finally transmit it to the edge computing module through ZigBee wireless communication.Users can use the mobile APP to get the historical electricity consumption data of different periods of time in the room.2.In order to predict the future electricity consumption trend in different periods,the deep learning algorithm is used to build a hybrid model,and the BLSTMAttention model is finally constructed to predict the future electricity consumption more accurately.3.Through experiments,the hardware system has the functions of efficient acquisition,stable transmission and data processing.The software system collects the historical electricity consumption information of each room,uses the deep learning algorithm model to predict the electricity consumption in different periods,and finally realizes the data visualization,so that users can intuitively observe the historical electricity consumption information and the future electricity consumption trend,providing a feasible scheme for the visual monitoring of household electricity consumption.