A fast Time-domain identification of bushing dynamics

• Published in: Mechanical systems and signal processing 2025-02, Vol.224, Article 112140
• Main Authors: Cosco, Francesco, Della Siega, Lorenzo, Adduci, Rocco, Gardonio, Paolo, Desmet, Wim, Mundo, Domenico
• Format: Article
• Language: English
• Subjects: Bushing identification; DMA testing; Time-domain parameter identification
• ISSN: 0888-3270
• DOI: 10.1016/j.ymssp.2024.112140

The identification of bushings dynamics is crucial for the development of innovative model-based digital tools, such as for example predictive maintenance or virtual sensing. Quite a few nonlinear modelling techniques have been proposed over the years, which can be suitably used to develop advanced models that capture the linear and nonlinear dynamics of a vast range of mechanical components and materials. Despite this, the experimental identification of bushings dynamics is usually accomplished with the so-called Dynamic Mechanical Analysis (DMA). This technique involves complex measurements based on several time-harmonic tests with different frequencies and amplitudes. This paper proposes a cost-effective time-domain technique for the identification of the non-linear dynamics of bushing elements. A single-test approach is proposed, which significantly cuts the measurement time, but also simplifies the measurement procedure. The proposed identification method is validated experimentally with respect to the frequency-domain DMA-based approach. The parameters of three models, having incremental complexity, are identified using both approaches. Also, three validation tests are used to quantify the accuracy of the proposed approach. The results presented in this paper indicate that, irrespective of the model complexity, the dynamic response of the bushing models identified using the proposed time-domain approach is generally more accurate than those built with the classical DMA-based procedures, particularly for simple models. •A Time-domain identification methodology is proposed for bushing components.•Compared to classical frequency-domain identifications it is faster and simpler.•It does not need cumbersome post-processing of measured data.•It generates more accurate parameters than classical frequency-domain identifications.