Abstract

AbstractThe response surface method (RSM) as a popular fitting method has been applied to numerous engineering issues. However, the existing RSM models cannot satisfy the analysis of complex nonconvex structure problems, especially in composite cylindrical shells. Thus a trigonometric mixed response surface method (TMRSM) is suggested to improve the accuracy and the fitting performance combing optimal Latin hypercube design (LHD), the moving least square method and the trigonometric functions. An improved robust design optimization (RDO) process is performed based on the weighted norm method and reliable constraints using the above TMRSM model to obtain more robust optimal design results. Three example functions and a composite cylindrical pressure shell are given to verify the effectiveness of the TMRSM model. This pressure shell design can check the robustness of the improved RDO process.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call