Surrogate Model Optimization Using Gaussian Process Regression and Particle Swarm Optimization for Engineering Design Problems
Article Sidebar
Main Article Content
Abstract
This research proposes surrogate model optimization (SMO) for engineering design by the use of machine learning regression algorithms and an optimization algorithm. The SMO proposed in this research features an optimization process designed to use only a small number of generated solutions. Additionally, variable scaling is applied prior to constructing the surrogate model to ensure that each variable has a similar impact on finding the optimal solution. The machine learning regression algorithms to be evaluated are gaussian process regression (GPR), support vector regression (SVR), and artificial neural network (ANN), while the optimization algorithm used is particle swarm optimization (PSO). After testing with benchmark function optimization problems, GPR with PSO has the highest performance and outperforms the design of experiment (DOE) technique, Taguchi method. For three standard benchmark functions, the proposed SMO using GPR and PSO yielded average errors of 0.0001, 0.0000, and 0.0856, whereas the Taguchi method resulted in average errors of 0.0000, 0.1249, and 1.0000, respectively. Thereafter, SMO using GPR and PSO can effectively solve engineering design benchmark problems and engineering design problems with the numerical calculation by Solidworks software. It produced design objective values with deviations of only 0.63% and 0.58% from the exact solutions for two standard engineering design problems where the exact solutions are known. Therefore, SMO using GPR combined with PSO is considerably suitable for surrogate model optimization in engineering design problems.
Article Details
References
H. Zhu, Z. Zhou, M. Wang, X. Yu, Y. Wu. C. Chen, and J. Qiao, “Integrated Design Optimization Method for Pavement Structure and Materials Based on Further Development of Finite Element and Particle Swarm Optimization Algorithm,” Construction and Building Materials, vol. 426, 136080, 2024.
X. Li., H. Su., J. Yang, G. Gao, and Y. Wang, “NURBS-boundary-based Quadtree Scaled Boundary Finite Element Method Study for Irregular Design Domain,” Engineering Analysis with Boundary Elements, vol. 159, pp. 418-433, February 2024.
S. Zhang, C. Deng, F. Wei, J. Li, X. Luo, and J. Ma, “Efficient Structure Optimization of Downhole Thermal Management System via Transfer Learning Surrogate Modeling,” Journal of Energy Storage, vol. 151, 120543, 2026.
R. Dai, J. Jie, Z. Wang, H. Zheng, and W. Wang, “Automated Surrogate-assisted Particle Swarm Optimizer With an Adaptive Parental Guidance Strategy for Expensive Engineering Optimization Problems,” Journal of Computational Design and Engineering, vol. 12, no. 3, pp.145-183, March 2025.
P. Manmatharasan, G. Bitsuamlak, and K. Grolinger, “AI-driven Design Optimization for Sustainable Buildings: A Systematic Review,” Energy and Buildings, vol. 332, 115440, April 2025.
C. Yang, J. Yang, and Y. Qin, “Research on Comparative of Multi-surrogate Models to Optimize Complex Truss Structures,” KSCE Journal of Civil Engineering, vol. 28, no. 6, pp. 2268-2278, 2024.
H. Xie, T. Wang, D. Meng, W. Zeng, and L. Guo, “Efficient Optimization of Noise-reducing Orifice Plates in Nature Gas Pressure Regulators Based on Adaptive Multi-scale Sampling-kriging Model,” Scientific Reports, vol. 16, 5872, 2026.
T. Kohno, N. Funayama, L. Xiao, S. Yamaguchi, and S. Imazato, “Predicting and Optimizing Viscosity of Dental Resin Composites With Gaussian Process Regression and Bayesian Optimization,” Dental Materials, vol. 42, no. 7, pp. 1112-1119, 2026.
J. J. Dang, S. Park, Y. Kim, “Design Optimization of Half-wave Resonator (Hwr) With a Machine Learning-based Framework Using Gaussian Process Regression and Multi-objective Genetic Algorithms,” Nuclear Engineering and Technology, vol. 58, no. 9, 104393, 2026.
Y. Yang, X. Song, Z. Deng, and J. Yu, “Gaussian Process Regression-Based Optimization of Fan-Shaped Film Cooling Holes on Concave Walls,” Fluid Dynamics and Materials Processing, vol. 22, no. 1, 9, 2026.
Z. Zheng, Q. Xu, Y. Feng, Y. Dai, Z. Chen, and Y. Wu, “Optimization Design of a Pre-swirl Duct for a Container Ship: A Simulation-based Design Method Integrating a Support Vector Regression Surrogate Model,” Ocean Engineering, vol. 352, part 1, 124574, 2026.
E. Mohammadipour, F. Adib, and M. Ghorbani, “Machine Learning–guided Optimization of PEO Coating Parameters Using Support Vector Regression for Enhanced Corrosion Resistance of AZ31 Magnesium Alloy,” Results in Engineering, vol. 29, 109408, 2026.
Z. Wang, C. Zheng, J. A. Mulyanto, Y. Wu, M. Chen, G. Yang, “Artificial Neural Network Surrogate Modeling and NSGA-II Algorithm-driven Aerodynamic Optimization for Wind-resilient and Sustainable Supertall Buildings,” Advanced Engineering Informatics, vol. 69, part D, 104047, 2026.
A. Allouhi, M. B. Amine, K. A. T. Aoul, “Metaheuristic Multi-objective Optimization With Artificial Neural Networks Surrogate Modeling for Optimal Energy-economic Performance for CSP Technology,” Energy and AI, vol. 20, 100488, 2025.
J. Kennedy, R. Eberhart, R. “Particle Swarm Optimization”. Proceedings of IEEE International Conference on Neural Networks, vol. 4, pp.1942–1948, 1995.
F. Ülker, S. Emiroglu, and Y. Uyaroglu, “A Combined Double Chaotic Maps and Subswarm Based Enhanced Particle Swarm Optimization for Global Optimization and Engineering Design Problems,” Alexandria Engineering Journal, vol. 142, pp. 91-113. April 2026.
K. Djermouni, S. Tamalouzt, A. Berboucha, K. F. Sayeh, L. Abad, Y. Belkhier, M. Benbouzid, “Improving Real-time Photovoltaic Power Delivery Through Integrated Advanced Fault Localization and Particle Swarm Optimization,” Applied Energy, vol. 414, 127786, 2026.
W. C. Weng, F. Yang, and A. Elsherbeni, “Electromagnetics and Antenna Optimization Using Taguchi’s Method,” Synthesis Lectures on Computational Electromagnetics, vol. 2, no 1, pp.1-94, 2007.
C. Lyu, Y. Song, D. Yang, W. Wang, S. Zhu, Y. Ge, L. Wang, “Surrogate Model of Liquid Cooling System for Lithium-Ion Battery Using Extreme Gradient Boosting,” Applied Thermal Engineering, vol. 213, 118675, 2022.
M. Matsunaga, C. Fujio, H. Ogawa, Y. Higa, and T. Handa, “Nozzle Design Optimization for Supersonic Wind Tunnel by Using Surrogate-Assisted Evolutionary Algorithms,” Aerospace Science and Technology, vol. 130, 107879, 2022.
L. Mustapha, H. Thorel, and Y,. E. Herpe, “Design of Experiments to Assess the Effect of Culture Components on Metabolism Expression and Albumin Secretion in Heparg Cell Line,” Pharmacological Research - Reports, vol. 5, 100084, 2026.
A. R. Parkinson, R. J. Balling, and J. D. Hedengren Optimization Methods for Engineering Design, Brigham Young University Press, 2013.
