Open Access

ARTICLE

Reliability Topology Optimization Based on Kriging-Assisted Level Set Function and Novel Dynamic Hybrid Particle Swarm Optimization Algorithm

Hang Zhou^*, Xiaojun Ding, Song Chen, Qijun Zhang

School of Computer Engineering, Chengdu Technological University, Chengdu, 611730, China

* Corresponding Author: Hang Zhou. Email:

(This article belongs to the Special Issue: Machine Learning-Assisted Structural Integrity Assessment and Design Optimization under Uncertainty)

Computer Modeling in Engineering & Sciences 2025, 144(2), 1907-1933. https://doi.org/10.32604/cmes.2025.069198

Received 17 June 2025; Accepted 31 July 2025; Issue published 31 August 2025

Abstract

Structural Reliability-Based Topology Optimization (RBTO), as an efficient design methodology, serves as a crucial means to ensure the development of modern engineering structures towards high performance, long service life, and high reliability. However, in practical design processes, topology optimization must not only account for the static performance of structures but also consider the impacts of various responses and uncertainties under complex dynamic conditions, which traditional methods often struggle accommodate. Therefore, this study proposes an RBTO framework based on a Kriging-assisted level set function and a novel Dynamic Hybrid Particle Swarm Optimization (DHPSO) algorithm. By leveraging the Kriging model as a surrogate, the high cost associated with repeatedly running finite element analysis processes is reduced, addressing the issue of minimizing structural compliance. Meanwhile, the DHPSO algorithm enables a better balance between the population’s developmental and exploratory capabilities, significantly accelerating convergence speed and enhancing global convergence performance. Finally, the proposed method is validated through three different structural examples, demonstrating its superior performance. Observed that the computational that, compared to the traditional Solid Isotropic Material with Penalization (SIMP) method, the proposed approach reduces the upper bound of structural compliance by approximately 30%. Additionally, the optimized results exhibit clear material interfaces without grayscale elements, and the stress concentration factor is reduced by approximately 42%. Consequently, the computational results from different examples verify the effectiveness and superiority of this study across various fields, achieving the goal of providing more precise optimization results within a shorter timeframe.

---

Keywords

---