Research on Magnetic Flux Leakage Signals Quantity Technology of Tank Floor Corrosion Defects Based on Artificial Neural Network

摘要

Magnetic flux leakage testing method is a major direction of tank floor testing. In this paper, the spatial distribution of magnetic flux leakage field of tank floor corrosion defects is analyzed based on the features of magnetic flux leakage signals. BP neural network model is applied to the quantity analysis of tank floor corrosion defects. The results in network training and test reach the quantitative accuracy requirements of tank floor corrosion defects, the established BP neural network is effective to the quantitative recognition of depth and width of the defects.