Optimal Design of Engineering Problems and Radial Basis Function Kernel Application

The aim of this paper is the application of Geometric programming (GP), and Lingo software for optimization of practical engineering problems and sensitivity analysis of such problems. The first part of the study is the basic definitions of the GP method and some rules for changing engineering problems to the standard form of GP. Sensitivity analysis was performed and to do this, the coefficients of the objective function and constraints were changed percent and the optimal response and optimal design parameters to these changes were evaluated. In the sensitivity, analysis for a series of changes in the coefficients, the optimal response to these changes will remain constant is very important . For validation and more evaluation, the results of the GP and Lingo software were compared with Lagrangian multipliers method (LMM). The result of this comparison shows the GP to be an efficient, powerful, and reliable method for optimizing linear and nonlinear problems and Lingo to be an easy and suitable program for finding optimal values and analyzing the solutions to engineering problems. The support vector machine (SVM) was used to predict vertical height (h), mean diameter and minimum weight of truss. After successful trials, the method predicted the abovementioned outputs with high precision.