Fuzzy Design Optimization-Based Fatigue Reliability Analysis of Welding Robots

In this work, a fuzzy design optimization-based fatigue reliability analysis technique is addressed to obtain the optimal fatigue reliability of the structures with random parameters under the minimum fluctuation of fatigue life. The challenge of the problem lies in uncertainties involved from both structural parameters and loads, which renders the fatigue reliability becoming the primary problem to be considered in evaluating structural performance. In order to obtain the optimal fatigue reliability, the optimal mathematical models aiming at the mass and fatigue life of welding robots are constructed respectively, and the genetic particle swarm optimization algorithm and radial basis function neural network (GAPSO-RBFNN) based surrogate model is then presented. Moreover, fuzzy constraints are introduced into the optimization model, and the improved non-dominated sorting genetic algorithm (INSGA-III) optimization strategy is proposed to solve it, which can effectively reduce the workload by increasing the diversity of Pareto solutions during the iterative process. Finally, the optimized structure combined with numerical simulation method is adopted to analyze the fatigue reliability. After analysis steps are given in detail, a practical welding robot structure is presented to demonstrate the validity and reasonability of the developed methodology.