Nonlinear Eddy Current Technique for Fatigue Detection and Classification in Martensitic Stainless-Steel Samples

The increasing use of stainless steel in industrial structures can be attributed to its excellent mechanical properties at elevated temperatures. Martensitic grade stainless-steel is used, for example, to manufacture steam turbine blades in power plants. The failure of these turbine blades can result in equipment damage contributing to expensive plant failures and safety concerns. Degradation and structural failure of these blades is largely attributed to material fatigue, at the microstructure level. Hence, it is important to evaluate the level of fatigue prior to the initiation of macro defects to ensure the viability of these components. Conventional nondestructive evaluation (NDE) techniques such as ultrasonic testing and eddy current testing are suitable in detection of macro defects such as cracks, but not very effective in evaluating degradation of the material at a microstructure scale. This article investigates the feasibility of the nonlinear eddy current (NLEC) technique to detect fatigue in martensitic grade stainless-steel samples along with a methodology to classify the samples. K-medoids clustering algorithm and genetic algorithm are used to classify the samples according to the severity of fatigue. Initial results indicate that stainless-steel samples, in different stages of fatigue, can be classified into broad categories of low, mid, and high levels of fatigue.