Application of artificial neural network in the construction of a lubricants selection system for sheet metal drawing

摘要

An intelligent system for the selection of lubricants in sheet metal drawing is constructed by using a model of artificial neural network that takes both the principal technological parameters and several parameters of lubricant into consideration. The average coefficient of friction under flange during drawing is predicted by the system and then selections are made to choose proper lubricant under certain operating condition according to it. The predicted results show good agreement with experimental data. The experimental setup designed by us is described concisely, and then a formula for correcting experimental errors is introduced which works to give more accurate data.