Time-variant Hybrid Reliability-based Design And Its Applications

The impact dynamic uncertainties play a more prominent part in the manufacturing process with the increasing complexity of industrial products.Influenced by the dynamic uncertainties,the performance of the products can fluctuate and the user security cannot be guaranteed.Considering the design optimization under uncertainties can provide not only an enhanced safety level but a better performance.The booming time-invariant and time-variant reliability-based design optimization methods have gradually been applied to practical engineering with the extensive attention of the probability-based uncertainties.However,some types of uncertainties cannot be modeled by probabilities,which brings challenges to traditional probability-based methods.The nonprobability-based hybrid reliability analysis and reliability-based design optimization methods should be further investigated.Therefore,based on the most flexible nonprobability theory,the interval theory,the present thesis will conduct the research on hybrid reliability analysis and design optimization.Research contents begin with the modeling of uncertainties,then the establishment of time-variant hybrid reliability optimization design model considering its dynamic,stochastic and non-probabilistic characteristics,finally the solutions to this kind of model and its applications.The research results of this thesis will solve the hybrid reliability optimization design problem with interval and probability uncertainty to a certain extent,and give designers flexible decision-making in the design process according to its interval characteristics.Based on this idea,the studies carried out in this thesis includes:(1)A complete time-variant hybrid reliability analysis framework is proposed.An efficient box moments reliability analysis approach which can effectively solve the upper and lower bound of the hybrid reliability is built.(2)Efficient methods for conservative and radical optimization problems are developed and a flexible-constrained reliability-based design optimization method is proposed according to the interval characteristic.The surrogate models are first used to solve the traditional conservative and radical optimization problems.For different preferences of designers,a flexible-constrained hybrid reliability-based design optimization method is further proposed.(3)An application of the proposed method to safety and energy saving-based design optimization of LEEX is carried out.The safety and energy-saving optimization design problem is solved,which fully proves the applicability of this method in complicated engineering problems.