Proved by long time prevention work and scientific research that coal and gas outburst takes place suddenly but has a lot of auguries including AE before outburst. Now, AE transmitting technology can't predict effectively because of a great deal of background noise and present noise processing technology, so the paper creatively proposed the method of noise processing based on wavelet packet transform and proved the feasibility of the method through a great deal of calculation and computer simulation. Based on basic principle of self-adapting neural network, the paper study the principle of AE self-adapting neural network forecasting coal and gas outburst in detail, and applied the method to forecast the danger of coal and gas outburst. It realized an AE dynamic trend prediction of the danger of coal and gas outburst. In the end, the paper designs the plan of the system simply.
|