Author(s)

Ji Lin, Quanliang Liu, Bo He, Jia Shao, Fulu Ji

Corresponding Author

Ji Lin

Abstract

This paper proposes a lightweight design method based on two-layer topology optimization of limited design space and parameter optimization to solve the lightweight problem of mechanical products better. In order to verify the effectiveness of the proposed approach, a lightweight design was carried out with the wall frame of a 70-meter ocean-going fishing boat trawl winch as an example. First, the conceptual model of the supporting structure is obtained through the two layers of topology optimization, and then the conceptual model is parameterized. After that, the optimal Latin hypercube design method is used to design the experimental structure changes to obtain multiple data points. Then, based on the obtained data points, the target performance of the supporting structure is sampled. Next, the radial basis function neural network based on the obtained sample points is established. Finally, the NLPQLP algorithm is used to solve the optimization model to obtain the optimal solution. The results show that applying the proposed method to the lightweight design of the support structure of a trawl winch can significantly reduce the quality of the support structure and get a very significant optimization effect while meeting the required target performance.

Keywords

Layer 2 topology optimization, Parameter optimization, Using nlpqlp algorithm