Defect reconstruction of submarine oil pipeline from MFL signals using genetic simulated annealing algorithm

This paper presents a genetic simulated annealing algorithm (GSAA)-based inverse algorithm for reconstructing the shape of a two-dimensional defect from the magnetic flux leakage (MFL) signals. In the algorithm, the GSAA formulated by incorporating the simulated annealing technique into the mutation operator of the standard genetic algorithm (GA) is employed to solve the optimization problem in the inverse problem, and a radial-basis function neural network (RBFNN) is utilized as the forward model. Experimental results demonstrated that the GSAA-based inverse algorithm is more accurate and is more robust to noise than the GA-based inverse algorithm.