Weight reduction of aluminum disc wheels under fatigue constraints using asequential neural network approximation method

This paper describes a weight reduction problem of aluminum disc wheels und er cornering fatigue constraints. It is a special structural optimization p roblem because of the existence of the implicit fatigue constraint. A seque ntial neural network approximation method is presented to solve this type o f discrete-variable engineering optimization problems. First a backpropagat ion neural network is trained to simulate the feasible domain formed by the implicit constraints using just a few training data. A search algorithm th en searches for the "optimal point" in the feasible domain simulated by the neural network. This new design point is checked against the true implicit constraints to see whether it is feasible, and the new training data is th en added to the training set. This process continues in an iterative manner until we get the same design point repeatedly and no new training point is generated. In each iteration, only one evaluation of the implicit constrai nts is needed to see whether the current design point is feasible. No preci se function value or sensitivity calculation is required. (C) 2001 Elsevier Science B.V. All rights reserved.