Fuzzy-logic application in the structural optimisation of a support plate for electrical accumulator in a motor vehicle

Abstract

In this paper, we analyse the design of a support plate for the accumulator of an electric motor vehicle. The support is an integral part of motor vehicle chassis. Therefore, geometrical configuration and boundary conditions require careful optimisation research of both function and structural behaviour, since lightness and dimension problems in the presence of dynamic stresses due to external factors have to be considered. Since these factors are complex and not homogeneous, the problem requires multi-criteria analysis. The presence of factors that are not precisely computable calls for fuzzy-logic application to optimisation problems, because fuzzy-logic is non-standard logic, particularly suitable for making choices in structural design. In plate optimisation, in fact, not numerically quantifiable characteristics such as a part’s workability, numerically determinable structural values such as stresses and strains, and analytically calculable properties such as weight come into play. These four parameters become the domain of fuzzy membership functions, by which we will extract membership grade values (co-domain). Design variables (domain) are plate thickness, ashlar’s number on the plate and stiffening ashlar’s depth. In our research, we characterise fuzzy correlation between parameters and required characteristics in order to determine, according to non-standard logics, the best topological configuration which corresponds to the optimisation of individualised characteristics in conformity with design constraints. Results show value improvement in stress and strain in comparison with the not yet optimised plate and small reduction in workability, whereas the mass is almost the same.