Crack Depth Profiling in Metallic Structures through ACFM Data using Type-II Fuzzy Logic System

One of the major issues in the industry is inhomogeneity depth profiling in the metallic structures before reaching them to the border of demolition. Fuzzy logic based methodologies, due to their ability to describe the complex issues with empirical nature such as non-destructive testing, are used for this purpose and usually provide acceptable results. But empirical rules and also extracted data from non-destructive testing methods mainly have high degree of uncertainty and therefore Classical fuzzy methods, which are based on exact membership grades and Type-I membership functions, are incapable to deal with them. Therefore, they cannot deal with noisy environments and also cannot represent a good performance for accurate depth estimation of unknown cracks. In this paper, to allocate uncertainty to rules and membership functions, the type-II fuzzy logic system is used to solve the inverse problem of crack profile depth estimation. Also Alternating Current Field Measurement (ACFM) signals are used to sizing the depth of crack profile. Then, experimental results of the proposed method are compared to the other state of the arts in the presence of different level of noise and different type of cracks. The results show the superiority of the proposed method to the other methods.