Application of modified subspace pursuit on guided wave NDT signal processing based on improved evolutionary algorithm

A kind of novel subspace pursuit method is proposed and applied to guided wave NDT signal decomposition and recognition to reduce the complexity of subspace pursuit. Modified differential evolution algorithm (MDEA) is applied to the modified subspace pursuit (MSP) by choosing chirplet function as match atoms. A steel pipe with hole and notch is detected by guided wave and the measured signal is decomposed and reconstructed by MSP. The matching result is compared to the matching result from OMP with evolutionary programming using mutations based on the t probability distribution (tEP). The defect echo can be identified easily from the matched signal. The matched parameters get by MSP and OMP are compared and analyzed. The defect locations are more exactly get from MSP than from OMP.