Robust topological optimization method of thin-shell structure considering thickness uncertainty

Abstract

The invention discloses a robustness topology optimization method of a thin-shell structure considering thickness uncertainty. The method comprises the following steps of firstly, determining the random field probability distribution of a thickness t; establishing a robustness topology optimization mathematical model of the structure; discretizing the random field by adopting an EOLE method to reduce the number of random variables; determining the sample points of the polynomial chaos expansion (PCE); calculating the structure response and the sensitivity information at the sample point based on a finite element method; calculating a mean value and a variance of the structure response; calculating an objective function value and the sensitivity information of the objective function; inputting the sensitivity and target function information into an MMA, and updating a design variable; and judging the result convergence, if the result is convergent, outputting an optimization result, and if the judgment result is not convergent, continuing iteration.

Description

technical field [0001] The invention relates to the field of optimization design of thin-shell structures of carriers and aircrafts, in particular to a robust topology optimization method for thin-shell structures. Background technique [0002] Carriers and aircraft structures generally adopt lightweight thin-walled structures. Due to plate raw materials, processing and manufacturing, etc., there is always a certain deviation between the thickness of the official product of the thin-walled structure and the design value, making the difference between the structural performance and the design value. There are variations between batches and batch-to-batch performance. The demand for refined design of aerospace vehicles and aircraft structural systems with strict quality requirements is also increasing. In the case of relatively fine thickness design, small changes will also cause large changes in structural performance and affect the reliability of the structure. . [0003] ...

AI Technical Summary

Problems solved by technology

[0004] (1) Designed according to the lower limit of the design thickness tolerance, it is easy to cause most of the actual thickness of the shell to be higher than the design value, causing the actual product to be too heavy and affecting the overall performance;

[0005] (2) Designing according to the nominal value will easily cause the actual local thickness to be too small, and the product reliability will be reduced. In severe cases, the structural product may be directly scrapped, seriously affecting the overall quality of the model, and resulting in an increase in the average cost of the product;

[0006] (3) The existing design method does not consider the local design parameter changes caused by the process, resulting in the actual separation between the design and the process, the design-process coupling design cannot be carried out, and the product design-production iteration cycle is long

Images