Title :
Prediction of Gas Emission Volume Based on Wavelet Neural Network
Author :
Gao, Li ; Hu, Yanjun ; Chen, Guizhen ; Yu, Hongzhen
Author_Institution :
Xuzhou Normal Univ., Xuzhou
Abstract :
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.
Keywords :
air pollution; environmental science computing; genetic algorithms; neural nets; wavelet transforms; Daubechies wavelet; backpropagation neural network; compact metric structure; gas emission volume prediction; genetic algorithm; nonlinear time series; wavelet neural network; Accidents; Accuracy; Continuous wavelet transforms; Control systems; Explosions; Genetic algorithms; Mathematical model; Monitoring; Neural networks; Wavelet analysis;
Conference_Titel :
Natural Computation, 2007. ICNC 2007. Third International Conference on
Conference_Location :
Haikou
Print_ISBN :
978-0-7695-2875-5
DOI :
10.1109/ICNC.2007.561
