Sign in Subscribe
Technical institutes

Adaptive Ensemble Surrogate Models for Mechanical Engineering Applications

Strategic Insights

— 1 min read

Researchers at Zhejiang Science Technical University in Hangzhou, China, have developed a novel approach to enhance the accuracy and efficiency of adaptive ensemble surrogate models (AESMs). The study proposes an adaptive ensemble surrogate model based on fuzzy C-means (FCM-AESM) and a parallel adaptive infilling (PAI) strategy to optimize new samplings and eliminate points in close proximity. The FCM-AESM demonstrates superior model prediction performance, and the ES-PAI strategy exhibits greater robustness and outperforms individual adaptive infilling (IAI) strategies.

Key Takeaways:

  • The FCM-AESM is initially proposed to enhance the accuracy of AESMs in practical engineering applications.
  • The FCM-AESM improves the modeling method for AESMs and its IAI strategy by applying FCM analysis to the initial training set.
  • The ES-PAI strategy optimizes new samplings and eliminates points in close proximity based on the Euclidean distance criterion.
  • The ES-PAI strategy consistently surpasses IAI strategies in both global and local performance and exhibits greater robustness.
  • The FCM-AESM method assisted by ES-PAI reduced the weight of a forklift gantry by approximately 22.18% while satisfying stress and deformation constraints.
  • The research has been peer-reviewed and implemented in the multiobjective optimization (MOOP) of a forklift gantry.
  • Authors of the research include Liye Lv, Tingxuan Yang, Yujun Lu, Wei Zhang, and Zhiwei Xiong.

Statistics:

  • The FCM-AESM reduced the weight of the forklift gantry by approximately 22.18%.
  • The ES-PAI strategy consistently outperforms IAI strategies in both global and local performance.
  • The research is implemented in the multiobjective optimization (MOOP) of a forklift gantry.

Sources:

  • An Adaptive Ensemble Surrogate Model Based On Fuzzy C-means and Its Parallel Infilling Strategy. Journal of Mechanical Design, 2025;147(11).
  • Asme, Two Park Ave, New York, NY 10016-5990, USA.
  • NewsRx. New Findings in Mechanical Science Described from Zhejiang Science Technical University (An Adaptive Ensemble Surrogate Model Based On Fuzzy C-means and Its Parallel Infilling Strategy). Mathematics Week. November 4, 2025; p 216.

Read more