Prediction of Gas Emission Volume Based on Wavelet Neural Network

An accurate prediction of gas emission volume under the shaft is the premise for the prevention of gas explosion. To more accurately predict the gas emission volume, a novel wavelet neural network is proposed in this paper. First, a new structure of wavelet neural network is established. This structure is a kind of compact metric structure, in which the Daubechies wavelet is adopted. And then genetic algorithm is used to study the weight of wavelet neural network. The simulation experiment shows the gas emission volume is non-linear time series. Finally, the experiment shows the predicted result made by wavelet neural network is more accurate than that made by BP neural network. Besides, this novel method can shorten the learning time and quicken the learning speed.