Evolutionary Strategy-Based Location Algorithm for High-Resolution Lamb Wave Defect Detection With Sparse Array

Intelligent defect location algorithms based on the times-of-flight (ToFs) of Lamb waves are attractive for nondestructive testing (NDT) and structural health monitoring (SHM) of structures with large geometric sizes. Unlike the classical imaging algorithm based on projecting the amplitude information of scattering signals into a discrete spatial grid on the structure via their propagation characteristics, intelligent defect location algorithms are more efficient in specific applications. In our previous work, an intelligent algorithm for the location of defects in plates was proposed by considering the statistical, diversity, and fuzzy characteristics of the classical defect location algorithm. This approach can realize the efficient location of different defects under a suitable parameter selection. However, interfering components remain in the results, which decreases the detection resolution. Because the measurement uncertainty is directly related to the time, an optimized intelligent location algorithm is provided for the efficient defect location with Lamb waves and a sparse transducer array in this study. The defect position is identified with high resolution by analyzing the distribution of individuals. Several specific data and a fuzzy control parameter are introduced to the proposed algorithm. The K-means algorithm was adopted to realize the adaptive updating of individuals. The influence of parameter values on the detection results was analyzed. A combined analysis of the individuals was provided to ensure the detection robustness by eliminating the influence of fuzzy control parameters on the detection. Compared with the elliptic imaging algorithm, the intelligent defect location algorithm has higher location resolution and executes approximately 65 times faster.