Multi-objective Optimization Of Vehicle Side Panel Cross Sections Based On Composite Materials

Sheets with hollow sections are widely used in vehicles,aircrafts and ships.On one hand,due to the massive usage of sheets,their mass is of a large proportion of the total mass of vehicles,thus its lightweight is of importance in reducing the total mass of vehicles.On the other hand,in consideration of aspects such as safety,satisfying the constraints such as stiffness and strength as well as reducing the mass is also of great importance.Thus we discussed the optimization problem of such sheets via theoretical derivation,engineering case study,and numerical prototypes,including the single objective topology optimization of carbon reinforced woven fabrics that leads to a lightweight result,a multi-objective optimization algorithm based on the improved hypervolume factor,and the multi-objective topology optimization method based on the proposed multi-objective optimization algorithm.Our work consists of the following parts:1.A component-based optimization method of component material sheets.Inspired by the moving morphable component method(MMC),we introduced a component-based topological optimization method for anisotropic carbon fiber woven fabrics with high manufacturability that explicitly describes the mechanical performance along with the geometric directions.We thus implemented our algorithm on the MATLAB platform,and verified the effectiveness on the practical case of the optimization of the hollow section of the vehicle side plates.The stiffness on the two major directions is no less than the initial design,and the mass reduce is up to 34.96%.The verification on Abaqus showed that our method has a good optimized performance.2.A multi-objective optimization method based on the improved hypervolume indicator.Aiming to construct the most competitive multi-objective solution set that is of necessity of practical engineering as well as to solve the problems that the gradient-based multi-objective optimization algorithm based on normal constraint method cannot set constraints under highdimensional situation and the weight-sum based multi-objective optimization method cannot achieve the concave Pareto front,we proposed a multi-objective optimization method based on the improved hypervolume factor,and gave an interpret of the construction of the objective function and its meaning,as well as its sensitivity analysis via theoretical derivation.Especially,we gave a detailed introduction of local upper bounds and the related problems.The superiority of the proposed method was claimed from two aspects of theory and experiment verification.3.Multi-objective topology optimization methods based on the improved hypervolume indicator.By combining the proposed multi-objective optimization algorithm and topology optimization method,we proposed two practical multi-objective topology optimization methods.Firstly,a bi-objective topology optimization problem for minimizing mass –compliance of MBB beam was conducted,and the result shows that the obtained optimal Pareto solutions have uniform distribution and other good performance.Moreover,we performed a multi-objective optimization of side plates made of carbon fiber reinforced woven fabrics,and thus led to multiple designs that satisfy different performance constraints.We classified the Pareto optimums via cluster analysis,and provided solutions of diversity for practical engineering.