Topology optimization introducing manual intervention

Abstract Topology optimization is an iterative process driven by algorithms from initial to final design, which can provide novel structural forms and is typically used to provide conceptual design. However, in real-life structural design, the intentions of designers, including requirements, specifications and various constraints that are difficult to modelling, are unable to directly introduced to the topology optimization algorithms. Therefore, it often takes a lot of time and even repeated iterations to modify the results of topology optimization, thereby extending the design process and increasing design costs. For this issue, in order to endow designers with the ability to modify intermediate results during iterations and integrate human design intentions with algorithms, this paper proposes a method called Manual Intervention. Our purpose is to provide a pause option for topology optimization process then modify the optimization intermediate results based on the expression of topology shapes (density, level set function, etc.), which is the literal explanation for manual intervention. An interactive interface based on radial basis function parameterized level set method was written on the MATLAB GUIDE platform. We have implemented a design method for interaction between manual intervention and algorithm meanwhile provided a simple case of maximizing stiffness. Furthermore extending the implement to SIMP. The final result shows that this method can effectively handle design requirements which are relatively complex and difficult to express through algorithms to integrate them into the design results. Its method has important potential and provides new ideas for future topology optimization design.