The Research of Acoustic Emission Signal Classification

When using acoustic emission (AE) monitoring of rock burst, the signals received by AE monitoring device are related to the type of AE source type, which influences predicting the AE monitoring accuracy. In view of the time-varying characteristics of acoustic emission signals, we adopt the Short-Time analysis technology, in other words, to acoustic emission signal transient analysis technique to extract the signal characteristics of effective, then the fisher criteria for the number of signal compression. Using neural network technology for signal classification, the classification results showed that this method is particularly effective in terms of the Acoustic Emission signal.