Based On Wavelet Transform And Neural Network Short-term Load Forecast

Accurate forecast of short-term electrical load is very important to the powersystem's security and economy. A new model is proposed which based on combining thewavelet transform and neural networks for load forecasting in this thesis. Someforecasting results are obtained for electrical load of Nanjing Area.By analyzing the electrical load, we find that the load curve shows certain regularity.Then using the Self-Organizing Map network (SOM), the load sequence of one week canbe divided into four load types. And by the good time-frequency characteristics of thewavelet transform, the load serial is firstly decomposed to different sub-serials using theMallat's pyramidal algorithm. Each sub-serial shows the different frequencycharacteristics of the load. Different artificial neural networks are constructed to predicteach periodical sub-serial according to their characteristics. The network of eachsub-serial mainly differs in selection of input variables of the network. To acceleratetraining neural network and to improve the convergence, an improved LM algorithm isadopted in artificial neural networks are used for each time interval (such as one net foreach hour). In addition, the methods of abnormal data processing based on wavelet theoryare presented dentally and simulated experimentally.The results of Nanjing load forecasting show that the WVNN method possesseshigher forecasting accuracy and better adaptability than artificial neural network(ANN)forecasting methods which considers day average and day type. For other time seriesforecasting problems, such as product price forecasting, the international crude oil priceforecasting, and so on, the method is also with high reference value and guidingsignificance.