Real-time identification of brittle material crack under high pressure based on AE signal analysis

Determined by the generating mechanism of AE event, AE signal has significant energy characteristics which can be used for identification. In this paper, a real-time embedded crack identification system of brittle material is designed and implemented. The system is fully functional that integrates AE signal acquisition, de-noising, signal extraction, characteristic recognition and alarm. In addition, an Energy-ratio based signal processing method is developed on the basis of statistical analysis of numerous brittle material crack data. Both the proposed system and method are verified on a test-bed deployed in a factory workshop. Experiment results show that the method has a satisfying identification accuracy and real-time performance in the noisy environment.