The Application Research Of Electric Power Load Forecasting Based On Time Series Data Mining Technology

Time series are very common in datasets. Time Series Data Mining has been one of the focuses of current data mining research. Recently the study on Time Series Data Mining mainly concentrates on both the similarity search in a time series database and the pattern mining from a time series. In the pattern mining, the trend prediction is a new domain. It mines the rules from time series data and makes use of the rules in predicting what will happen in the future. This dissertation combines time series analysis technology with data mining theory, researches a new data mining method based on time series analysis technology which can be used on forecasting the future data. This method can build an input sample pattern which is suit for the BP Neural Networks of data mining and finally find the law of the system by studying from time series again and again. At the same time, to counter the original BP algorithm deficiency, the writer made a further improve on that. The electric power system's live recording data is the research object of this text. Proceed from the local actual conditions of Fuxin electrified wire netting in motion, this text developed the relevant application software of electric power load forecasting. The simulated test result indicates the rationalization of the pattern builded in this dissertation and it can greatly raise the accuracy degree of electric power load forecasting. This dissertation have done helpful exploration on popularizing data mining theory based on time series analysis technology.