Radial Basis Functions for Topology Optimisation of Engineering Structures (June 2026).

This presentation was given by Edgar Buchanan at the 2026 IEEE World Congress on Computational Intelligence (WCCI)  the world’s leading event for Computational Intelligence research and applications (21–26 June 2026) in Maastricht.

Abstract: Truss topology optimisation is a well-established field in structural engineering, traditionally focusing on determining optimal node locations and member sizing to minimise compliance or weight under prescribed loads. Recent advances have explored more flexible design representations to overcome limitations of fixed-topology approaches. In this context, Radial Basis Functions (RBFs) offer a tool for topology optimisation, providing smooth, continuous control over geometry and structural parameters. By incorporating RBFs, optimisation algorithms can efficiently generate diverse structural solutions, enable finer control over cross-sectional areas, and improve convergence rates. This work highlights the potential of RBF-based representations to enhance truss topology optimisation, offering designers a versatile and computationally efficient framework for exploring complex, high-performance structural configurations. The optimisation of a car chassis subjected to multiple load cases is used as a case study to demonstrate the capabilities of the approach.
Index Terms—topology optimisation, evolutionary computation, radial basis functions, engineering design.