City Research Online

Abstract

This study presents a novel approach for estimating the damping coefficient and reaction force of screw based regenerative shock absorbers (RSAs) using a similarity-based recommender system. A detailed mechanical model of a screw-driven RSA was developed, and a dataset of ten training and three test cases was generated using variations in generator parameters and dynamic response metrics such as the Tratio. A recommender system utilizing an exponential similarity function was employed to infer damping coefficients and reaction forces for new test data based on similarities with training samples. The results demonstrate strong predictive performance, with damping coefficient estimation errors typically below 4% and reaction force predictions showing reasonable accuracy, though slightly more variable. The proposed framework offers a fast and accurate alternative to traditional parameter identification methods, with potential for real-time applications. The study also highlights the benefit of expanding the training dataset to improve similarity matching and reduce estimation errors in future implementations.

Publication Type:	Article
Additional Information:	Published by Elsevier Ltd. © 2026. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/
Publisher Keywords:	Screw Based Regenerative Shock Absorber (RSA); Recommender System; Time Ratio (Tratio); Reaction Force; Damping Coefficient
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Departments:	School of Science & Technology School of Science & Technology > Department of Engineering
SWORD Depositor:	Symplectic Administrator

Export

Downloads