Structure robustness optimization design method containing interval parameter uncertainty

Abstract

The invention discloses a structure robustness optimization design method containing interval parameter uncertainty. The method comprises the following steps that a structure robustness optimization design model based on an internal is built; sample points are obtained by adopting a Latin hypercube sampling and co-simulating technique; a Kriging proxy model for predicting a target function and a constraint function is constructed; an interval robustness optimization design model is solved by adopting a double layer-nested genetic algorithm, left boundaries and right boundaries of the target function and the constraint function are calculated in the inner layer of the genetic algorithm, and in the outer layer of the genetic algorithm, total interval constraint violation degree vectors of all design vectors are calculated, and the feasibility of the total interval constraint violation degree vectors is judged; all the design vectors are subjected to advantage and disadvantage sorting according to a superior relationship criterion based on the interval constraint violation degree vectors; when a largest evolution algebra or convergence threshold value is achieved, the optimal solution of the robustness optimization design model is output, and the structure robustness optimization design containing interval parameter uncertainty is achieved.

Description

technical field [0001] The invention relates to a robust optimization design method for an uncertainty structure with interval parameters. Background technique [0002] There are many uncertain factors in the optimal design of structures, such as fluctuations in material properties, manufacturing and assembly errors, and changes in the load environment. The design results obtained by the deterministic optimal design method ignoring these uncertain factors are often very sensitive to uncertain factors, and there are large fluctuations in the target performance of the structure, which cannot meet the work requirements. Therefore, the influence of uncertain factors must be considered in the process of structural optimal design . Since it is often difficult to obtain accurate probability distribution information of uncertainty factors in actual engineering, but only the variation range of uncertainty parameters, it is a simple and effective method to describe these uncertain fa...

AI Technical Summary

Problems solved by technology

Although this method has strong flexibility and a wide range of applications, there are problems such as excessive calculation

Dong Rongmei proposed a non-probabilistic robust optimization design method based on interval analysis in her doctoral thesis "Research on Robust Optimal Design Theory and Method for Engineering Uncertainty Problems". This method uses interval numbers to describe the uncertain factors in engineering practice. It is necessary to obtain its probability distribution in advance, but when dealing with multi-objective problems, it is still converted into a single-objective problem by weighting

In recent years, some scholars have proposed to use the direct method to solve the interval robust optimization design model of the structure, so as to avoid the introduction of transformation parameters and the loss of information in the transformation process, and greatly simplify the transformation process. However, the existing interval model direct solution method The obtained optimization results may not fully meet the requirements of the constraint function, that is, the robustness of the constraint conditions cannot be guaranteed. Therefore, in order to realize the robust optimization design of the structure with interval parameter uncertainty, it is necessary to study the robustness of both the objective and the constraint. A Direct Solving Method for Interval Optimal Models

Images