An Automated Design Method for Plane Trusses Based on User Preference Information

Structural design is a complicated decision-making process involving multiple qualitative and quantitative factors. Currently, most automated methods consider only objectives constraints, ignoring the information that difficult define mathematically, such as user preference for structural shapes. This limits functionality efficiency of methods. In this study, method named STSA-P proposed plane trusses to incorporate into automatic process. Two main problems are addressed, i.e., how quantify coordinate it with other objectives. A prediction model developed first problem by generating data set selecting an appropriate machine learning (ML) algorithm. Specifically, truss features quantitatively representing shapes identified sample population. Furthermore, interactive system collecting evaluation labels. Strategies reducing fatigue also considered during numerical experiments conducted select suitable ML Regarding second problem, physical programming modified construct new aggregate function which effectively coordinates cost designed considering constraints. On basis, incorporated Topology Shape Annealing (STSA) form method. students invited perform case using It demonstrated results verify practicality validity