Meta Modelling of Submerged-Arc Welding Design based on Fuzzy Algorithm

[摘要] Fuzzy algorithm based meta-model is proposed for approximating submerged-arc weld design factors such as weld speed and weld output. Orthogonal array design based on the submerged-arc weld numerical analysis is applied to the proposed approach. The nonlinear finite element analysis is carried out to simulate the submerged-arc weld numerical analysis using thermo-mechanical and temperature-dependent material properties for general mild steel. The proposed meta-model based on fuzzy algorithm design is generated with triangle membership functions and fuzzy if-then rules using training data obtained from the Taguchi orthogonal array design data. The aim of proposed approach is to develop a fuzzy meta-model to effectively approximate the optimized submerged-arc weld factors. To validate the meta-model, the results obtained from the fuzzy meta-model are compared to the best cases from the Taguchi orthogonal array.

[发布机构] Dept. of Naval Architecture and Ocean Engineering, Mokpo National University, Jeonnam, Korea, Republic of^1;Myung Il Co., Ltd., Jeonbuk, Korea, Republic of^2

[效力级别] 无线电电子学

[关键词] Fuzzy algorithms;Fuzzy if-then rules;Meta-modelling;Non-linear finite-element analysis;Orthogonal array designs;Taguchi orthogonal arrays;Temperature-dependent material properties;Thermo-mechanical