Reliability-based Design Optimization Of Crane Metal Structure Based On Constrained Boundary Sampling

Cranes are indispensable lifting equipment in modern manufacturing and logistics,and are widely used in manufacturing plants,mines,and large ports.The bridge system mainly plays the role of transmitting and bearing loads,and is the most critical subsystem of the crane.For crane design,the safety and reliability of the bridge design must be ensured to ensure the safety of the whole machine during use.Traditional crane design is mainly based on a deterministic design method based on semi-theoretical and semi-empirical formulas.On the one hand,this method has large design redundancy,cumbersome calculation steps,and low portability and substitutability;on the other hand,the design does not consider the crane’s due to the influence of uncertain factors in the manufacturing,installation and use process,reliability analysis and structural design are not combined.With the rapid development of computer-aided design methods,researchers and engineering designers have introduced finite element analysis technology into the crane design,and fully considered the reliability problems caused by uncertain factors,based on the finite element analysis of the crane metal structure(CMS)is designed for reliability optimization.However,the computational cost of this method is huge and the efficiency is low.Therefore,how to reduce the calculation cost and improve the efficiency while taking into account the accuracy of the calculation results has become the research focus of researchers.Using the Kriging surrogate models instead of the finite element model to optimize the reliability of the crane metal structure can reduce the calculation cost and also take into account the accuracy of the calculation results.However,the reliability optimization design of the crane metal structure of the Kriging surrogate models in the whole region still needs more high-fidelity samples to ensure the correctness of the optimization results,which causes the problems of high calculation cost and low calculation efficiency.Regarding the issue above,A RBDO of the CMS based on Kriging surrogate models of the constraint boundary sampling(CBS)method is proposed in this paper.According to the sampling rules,only more high-fidelity samples are drawn in the constraint boundary area,which improves the fitting accuracy of the Kriging surrogate models in the constrained boundary area and reduce the time consumption resulted by pursing the constructions of the high-precision surrogate models in unconstrained boundary regions.Verification by a case:Compared with the RBDO based on Kriging surrogate model in the whole region,the CMS’RBDO based on the CBS’ Kriging surrogate models could save about 78% of the calculation cost,when the same optimization results are obtained.It can be seen that the CBS method can greatly improve the design efficiency,meanwhile maintaining good accuracy in the RBDO.