Optimum Design of Truss Structures using Neural Network

The purpose of this paper is to verify the applicability of the prototype of the Neural Network base model for Optimum Structural Design(NNOSD) to search the optimum design in all cases immediately. NNOSD suggest the procedure of the optimum design and the proper values of the neural network variables(training pattern, number of hidden neurons, maximum training error, steepness coefficient, learning constant, momentum term, etc.) to reduce the computer time and to improve the accuracy. And the methods of the connection between NNOSD and the existing design system is presented. The artificial neural network is used as an alternative design model of the warren truss since it can handle uncertainty through the probability method.