After briefly describing the current research situation of electric power load forecasting, the dissertation firstly discusses the architecture model of electric power load forecasting system. And then, the dissolution focuses its emphasis on the research of electric power load forecasting based on data mining technology. Main work of this research consists of two parts:Part I , including chapter 2, studies the architecture model of electric power load forecasting system. By analyzing the current architecture model, we put forward on anew and universal architecture model-an architecture model of electric powerload forecasting system based on data mining technology.Part II is composed of chapter 3 to chapter 6. We mainly discuss electric power load forecasting based on data mining technology.Firstly, in chapter 3, we mainly discuss the question of the selection of key attributes in load forecasting model-building. We put forward on a method of mining best attribute set using information entropy.Secondly, in chapter 4, by the analyzing the advantages and the key points of using artificial neural network model for electric power load forecasting, we put forward on a method of electric power load forecasting based on fuzzy genetic neural network.Thirdly, in chapter 5, to solve the main problem of electric power load forecasting based on expert system, a method of electric power load forecasting based on fuzzy association rules mining is put forward on.In chapter 6, research results of this dissertation are summarized, and some directions for further research are also provided.
|