Research Of Electromagnetic Nondestructive Testing Based On Neural Networks

Metal material hold a very important part in industrial materials, parts made of which are widely used in industry. Thus, metal parts' quality must be inspected before using, can be used in the inspection of metal parts' quality. Electronmagnetic test, a nondestructive testing of metal material. Because there are many factors which have effect on Electronmagnetic test, the qualitative analysis is always a difficult problem.To settle the above problems, we ameliorated the electromagnetic inspecting instrument and develop neural network system, because neural network has self studying ability, self organizing memory and functing inferring ,it has the ability to do great deal of information-dealing work that other means can't realize. The application of neural network breaks away the traditional mode and provides us with efficient ideal and ways.On the base of the improved electromagnetic inspecting instrument, we design and develop the neural network system and improve the algorithm to raise the reliability and settle the shortcomings of LVQ neural network system. At the same time, the improved system can be easily used and the course is greatly simplified and the result is much more credible, so it is efficient to improve the traditional aspect that we can do little to deal with plenty of electromagnetic inspecting data.