The study presents a novel source location method based on EEMD （ensemble empirical mode decomposition） and optimized FastICA （independent component analysis） for determining the position of the AE （acoustic emis...The study presents a novel source location method based on EEMD （ensemble empirical mode decomposition） and optimized FastICA （independent component analysis） for determining the position of the AE （acoustic emission） sources in the damage structure of crane. Firstly, The AE signals are self-adaptive decomposed into a number of IMFs （intrinsic mode functions） by using EEMD algorithm. Then, the main feature IMFs signals are extracted as the effective AE source signal by optimized Fast-ICA method, the PSD （power spectral density） of each IMF and the real IMFs are obtained. According to the principle of spectrum similarity, time delay is computed at the different channels in combination with the Cross Correlation time delay estimation principle. Thirdly, a novel source location algorithm based on EEMD-FastICA is proposed and the results of AE source location are obtained. Finally, the three-point bending test for the crane is implemented in order to validate the efficiency of the proposed method. The experimental results indicate that the proposed method has the ability to determinate the position of the AE sources and reduce the interference noises. Moreover, compared with the traditional location algorithm, a considerable improvement is obtained.展开更多
基金Acknowledgments This work is supported by the National Natural Science Foundation of China （51575101, 51305176）, the funds of High-level talents of JIT （JIT6201623, FHXM201608）.
文摘The study presents a novel source location method based on EEMD （ensemble empirical mode decomposition） and optimized FastICA （independent component analysis） for determining the position of the AE （acoustic emission） sources in the damage structure of crane. Firstly, The AE signals are self-adaptive decomposed into a number of IMFs （intrinsic mode functions） by using EEMD algorithm. Then, the main feature IMFs signals are extracted as the effective AE source signal by optimized Fast-ICA method, the PSD （power spectral density） of each IMF and the real IMFs are obtained. According to the principle of spectrum similarity, time delay is computed at the different channels in combination with the Cross Correlation time delay estimation principle. Thirdly, a novel source location algorithm based on EEMD-FastICA is proposed and the results of AE source location are obtained. Finally, the three-point bending test for the crane is implemented in order to validate the efficiency of the proposed method. The experimental results indicate that the proposed method has the ability to determinate the position of the AE sources and reduce the interference noises. Moreover, compared with the traditional location algorithm, a considerable improvement is obtained.
