Detection of ring gear surface defects of wheel speed sensor based on neural network

Abstract

In view of the low efficiency and slow speed caused by the manual inspection of the wheel speed sensor ring gear in the automotive parts industry, this paper proposes a method for detecting the surface defects of the ring gear of the wheel speed sensor based on the neural network. The method is based on the single hidden layer BP neural network model, and the LM algorithm is used to train the network to achieve stability, the defect types are identified by combining the image feature parameters of various gear ring surface defects. The surface defects detection results of the wheel speed sensor ring gear show that the defects classification accuracy of this method is more than 94%, and the detection time of each ring gear is less than 4s. The detection result is better than the manual visual method.