Rapid optimization design of plate-type lifting lugs using an improved backpropagation neural network

Read full text Summarise with VIRAT
Authors: Mingliang Yang, Chao Wang, XiJun Xu, Heng Yang, Qing Dong, Keyuan Zhao
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Aiming at the common challenges in the traditional design of plate-type lifting lugs, such as structural redundancy and low iteration efficiency, this paper proposes a rapid lightweight optimization design method based on an improved backpropagation neural network. With the goal of structural lightweighting, a hybrid neural network model integrating particle swarm optimization and genetic algorithm is constructed. Leveraging its powerful nonlinear mapping capability, the model replaces traditional finite element trial calculations and achieves fast optimization of design parameters. A validation through seven practical engineering cases demonstrates that, under the premise of satisfying the same safety factors and working condition constraints, the proposed method significantly outperforms traditional design schemes, achieving a structural weight reduction of over 30% in all cases. Compared with that of the traditional particle swarm optimization algorithm, the optimization effectiveness of the proposed method is improved by approximately 14%. The results confirm the validity and reliability of this approach, providing a new technical pathway for the rapid optimization design of non-standard load-bearing components.

Fields of Study & Subjects

Metaheuristic Optimization Algorithms Research Engineering Applied Research Topology Optimization in Engineering

Publication details

Journal / source
Systems Science & Control Engineering
Volume
14
Issue
1
Publication date
20 March 2026
Language
EN