Research On The Reliability Optimization Methods Of Welding Technology Based On The Probability Of Defects

With the development of the global industrial modernization, welding technology as a kind of advanced manufacturing technology get the comprehensive innovation and pioneering engineering application, and play an important role in the national economy. However, welding process is a drastic heat treatment process, welding products easy to appear cracks, pores, incomplete fusion and residual stress, which make the bearing capacity of welding products be weakened severely, and bring greatly potential safety hazard to the equipment. Welding technology is the core of welding technology, which determines the welding quality of products. Therefore, how to effectively prevent and control the reliability of the welding technology is becoming more and more important.In this paper, according to the existing expression between weld defects and welding technology parameters, establish a multi-objective reliability optimization model, in which the cracks and pores are the objective functions, the probability of incomplete fusion and residual stress are the constraints. While minimizing cracks and pores, ensure that the probability of incomplete fusion and residual stress within a certain range. In order to better conform to the actual situation, the design variables are random variables in the model, and obey the known normal distribution. In order to solve the model, first of all, using the sequential optimization and reliability assessment method (SORA) transforms the model into optimization and reliability analysis, which is a single cycle sequence structure. Then, using the performance measure approach (PMA) converts reliability constraints to equivalent deterministic constraints. Secondly, adopting physical programming and fast elitist non-dominated sorting genetic algorithm for multi-objective optimization (NSGA-II) to solve the converted multi-objective optimization problem, get the current optimal solution. Finally, analyzing the reliability of the constraints according to the optimal solution obtained by deterministic optimization, get the most probable point (MPP), and judge convergence for MPP, if yes, end of the cycle, otherwise, move the equivalent deterministic constraints boundary to probability constraints boundary according to the optimal solution and the MPP, continue to optimize until convergence.This paper optimizes the welding technology parameters, effectively reduces the failure probability of welding products and improves the reliability of the welding technology. It can be used for quantitative analysis of welding technology reliability design, technology design scheme comparison and guide technique designer to conduct reliability optimization.