Property Identification During Intellectural Deep Drawing Based On Genetic Algorithm And Neural Network

The intellectualization of sheet metal forming is the process in which, through an organic combination of sheet metal forming theory with control theory and computer science, and based on the characteristics of the workpiece to be worked, the material properties are identified and the optimum technique parameters are predicted on-line by means of easily-monitored physical quantities and the forming is automated to those parameters thus determined.The intellectualization of axisymmetrical workpiece is an important research field of sheet-metal-forming intellectualization and is the basis of intellectualization of complicated curved face workpiece. The identification of material properties and coefficient of friction is one important part of intellectualization. The precision and the time of identification will affect the realization of intellectualization directly. Artificial neural network (ANN), which is a rapidly developing crossing subject, has many advantages such as parallel storing, parallel processing of information, teaching itself etc. Genetic algorithm (GA), a probability search algorithm, simulates the mechanism of natural choose and natural genetic and its main features are group research strategy and individual information exchange which don't depend on gradient information. ANN and GA have a broad application in information processing, pattern recognition, intellectual control and some other aspects. This thesis realizes the real-time identification of material properties and coefficient of friction by using the features and virtues of ANN and GA.This thesis establishes a GA-ENN model for identification of material properties and coefficient of friction during axisymmetric workpiece deep drawing intellectualization, makes the program of GA-ENN model using visual program language, integrates the intellectualization control system, and makes a study of the relative problems on the GA-ENN model. The real-time identification of material and coefficient of friction is successful, which is shown by the experimental results of different sheet metals from intellectualized deep drawing. It lays foundation for further study ofidentification of material properties and it is significant for further study ofintellectualization during complex curved parts deep drawing.