Power Load Forecasting Based on a Fuzzy-RBF Neutral Network

摘要

The author proposes a fuzzy RBF artificial neutral network-based load forecasting model ,also can improve BP's slow convergence velocity and being apt to get into partial minimum. This model applies new adjacent clustering algorithm in RBF network model so as to realize the adjustment of both network structure and parameters. It can effectively improve the training speed and forecasting precision. The proposed method has been implemented in an emulated forecast, and the result improves that the model possesses far superior forecast precision and speed, and also favorable stability and adaptability.