Effect of crack orientation in crack detection of a pipeline based on Eddy current testing and inverse problems using Grey Wolf Optimiser

The orientation of cracks is an essential aspect in the research of fracture behavior, particularly the investigation of crack growth and propagation in structures under fatigue or static and dynamic stress within a gas pipeline. This study used the Grey Wolf Optimizer (GWO) to improve the design of non-destructive Eddy current evaluation devices. In this paper, a numerical approach was developed based on the finite elements method (FEM) and an iterative inversion technique by GWO that minimizes a non-linear least squares objective function that assesses the overall discrepancy between forecasts and observations in a 3D gas pipeline in order to determine the orientation of cracks from Eddy current data. The idea of a GWO algorithm that looks for the best solution for a particular crack orientation problem within a broad solution space is explored. The method begins its search anywhere in the solution space and then searches the "local" region around the start point for better solutions; the process terminates when the algorithm determines what it thinks is the best solution or when preset stopping criteria are satisfied. As compared to existing meta-heuristics approaches, this novel approach provides a competitive performance.