Collaborative Design Method for Product Quality Based on Neural Network

The quality of plastic products, especially the surface quality, is closely related to the process parameters. This paper proposes a method for the classification and identification of surface defects, extracts the defect features from the surface image of plastic products to evaluate the surface quality, builds a mapping model of surface quality and process parameters based on BP neural network, fits the distribution of main process parameters using Gaussian mixture model, and proposes an update method of process parameters based on the gradient amplitude of the historical curve, which is helpful for the online control of product quality.