AE detection of crack signal in tank shell using the multi-sensors with adaptive weighted fusion method

In order to detect the crack accurately and make full use of the information from multi-sensors, this paper proposes a detection method based on the adaptive weighted fusion method and Kalman filter method for tank crack detection. The fusion method of four sensors is deduced, and the simulation with noise interference is utilized to analyze the effectiveness of the proposed method. An experiment is implemented to obtain the signals with noise interference from four sensors. The proposed method is used to obtain the weight of each sensor, the signals from four sensors are fused by the calculated weights, and then Kalman filter method is employed to remove the noise. Compared to the method with only Kalman filter method, the proposed method with adaptive weighted fusion has a higher signal noise ratio which is 9.89dB. The result shows that the proposed method can make full use of multi-sensor information and has a better noise suppression.