A sort of new energy consumption forecast method

A model of heredity nerve network with energy consumption forecast is proposed in this paper, which performs the multi-wavelets filter processing to the energy consumption value. Heredity algorithm is applied to optimize the initial weight value and the threshold value of the nerve network, and to overcome such inherent flaws as slow BP network training speed, which makes it easy to converge partial minimum point and weak overall search ability, etc. and enhance the forecast result accuracy to a certain extent. This text simulates the algorithm based on some steel companies. The result indicates that the model has higher forecast precision and quicker convergence speed compared with the traditional BP algorithm.