PCA-IBP model application in medium and long-term runoff forecasting

This paper proposes a med-long term runoff forecasting model based on the principal component analysis (PCA) and the improved BP Neural Network. PCA was utilized to eliminate the relevance between input data, reducing input dimension and effectively reducing the model´s structural complexity, improving the model´s learning efficiency and forecast performance. The proposed model was predicted and verified with actual data, compared and analyzed with the traditional BP neural network model. The results show that the proposed model is superior to the traditional BP neural network model in terms of forecasting efficiency and accuracy.