The appearance and growth of the microcracks in the structure is an important factor that influences the structure safety and its service life. Thus it is very important to detect the crack and monitor its growth at the beginning of the crack. Aiming at the main style of failures in metal structure - fatigue fracture, this paper research acoustic emission waveforms analysis that base on wavelet packets feature extraction, through processing acoustic emission signal to test metal fatigue fracture. First, this paper analyses the reason of metal fatigue fracture and introduces the theory of acoustic emission. Based on that, we establish the time domain module of acoustic emission signal and extract the feature of acoustic emission signal using wavelet packets. According to the experimental results bending specimen, acoustic emission techniques monitoring fatigue crack propagation is certificated not only to resemble variable rule of fatigue crack propagation but also to catch generation of fatigue crack in real time. Compared with the method of parameter extraction, this method can not only realize real-time and dynamic monitoring, but also get the result that is similar with fatigue crack expanding rate curve.
Improving of Acoustic Emission Signal Detection for Fatigue Fracture Monitoring