Research On Household Intelligent Power Control Strategy

Since the concept of smart grid was put forward,smart grid projects have been vigorously developed at home and abroad,especially in recent years,China’s smart grid and related industries have developed rapidly.For smart grids,accurate power forecasting and reasonable power dispatch are essential.The home user is one of the main power terminals of the smart grid,and it is also an important part of the information feedback of the entire grid system.Therefore,the forecast and dispatch of home power consumption is the basis for improving the efficiency and stability of the smart grid.Reasonable model establishment and selection of forecasting methods largely determine the quality of power consumption forecasting and dispatching,so it is necessary to study the accuracy of power consumption forecasting and the scientificity of power dispatching.The purpose of this thesis is to establish a reasonable and scientific scheduling model.At the same time,a BP neural network and an improved particle swarm optimization algorithm are combined and introduced into the field of power consumption prediction and scheduling.This thesis aims at the accuracy of power consumption prediction and user satisfaction after scheduling,and studies the effectiveness and scientificity of the above-mentioned hybrid artificial intelligence algorithm in power consumption prediction and scheduling.The main content of the thesis is as follows:Firstly,based on the actual data,the residents’ electricity consumption in China and the importance of electricity forecasting and dispatching were analyzed.The literature analyzed the advantages and disadvantages of various mathematical models and intelligent optimization algorithms.On this basis,a hybrid intelligent algorithm combining BP neural network and improved particle swarm optimization algorithm is selected for household electricity consumption forecast and household appliance scheduling.Secondly,the structure of a classic smart home energy management system is designed,and the functions and sub-modules of a conventional smart home energy management system are proposed.The basic concept and functional architecture of the smart home energy management system are explained,and each sub-module is introduced one by one,which includes the smart home energy management system center,smart meters,communication and network systems,and other smart devicesThen,based on the structure of the smart home energy management system and its application purpose,a scientific and reasonable mathematical model was established with the goals of saving electricity purchase costs and increasing user satisfaction.At the same time,based on the rationality of algorithm selection,BP neural network is introduced,and its algorithm structure and learning process are outlined.In order to improve the shortcoming of BP neural network easily falling into local extremum,the effectiveness of particle swarm optimization for its improvement is studied.Based on this,a particle swarm optimization algorithm based on the non-linear weight change of the natural constant e is further proposed,and the effectiveness of this improved algorithm is proved using standard test functions.Then,an improved particle swarm optimization algorithm was used to optimize the BP neural network,and a power intelligent prediction and scheduling model based on the hybrid intelligent algorithm was established.Based on the actual data of household electricity consumption at the current stage,software simulation studies and comparisons are conducted.The results show that the control strategy proposed in this thesis has high accuracy in electricity consumption prediction and is scientific and reasonable in electricity consumption scheduling.Finally,it summarizes the workload of the full text,and puts forward the work that needs to be further improved and studied in this thesis.It also points out the problems that need further research and mutual verification in the theoretical and practical applications of power consumption prediction and scheduling.Figure[27]Table[13]Reference[74]...