Intelligent Monitoring and Prognostics of Automotive Clutches (Intelligente bewaking en prognose van koppelingen voor voertuigen)

Wet friction clutches are machine elements enabling the power transmission from an input shaft to an output shaft during the operation, based on the friction occurring in lubricated contacting surfaces. These components play a critical role in vehicles equipped with automatic transmissions, power shift transmissions and limited slip differentials. An unexpected failure occurring in clutches can lead to an unexpected total breakdown of the vehicle. This undesirable situation can put human safety at risk, possibly cause long-term vehicle down times, and eventually result in high maintenance costs. In order to minimize the negative impacts, an optimal maintenance strategy, i.e. a condition based maintenance (CBM), driven by an accurate condition monitoring and prognostics therefore needs to be developed and implemented for wet friction clutches. A literature study has been carried out during this work in order to establish the state-of-the-art development of clutch monitoring and prognostics systems. Despite the fact that the performance demands for clutches increase, research in this direction is apparently very limited. Several methods related to this aspect have been reported in literature, but, they are not sufficiently pragmatic and flexible to implement for real-life applications. To fill this substantial gap, development of a clutch monitoring and prognostics method that can be easily implemented on real-life applications is the main objective of this thesis. There are two main failure modes that can interrupt the functionality of wet friction clutches as reported in the literature, namely (i) lubricant and (ii) friction material degradations. This thesis mainly focuses on the failure of wet friction clutches caused by the friction material degradation, which is a valid assumption for heavy-duty applications. In this scope, a better understanding of how friction material degradation affects the clutch dynamic behavior during its operation is of paramount importance for developing an effective clutch monitoring and prognostics system. In the first place, this thesis contributes to the contact modeling of nominally flat rough surfaces with asymmetric asperity heights distribution, which is relevant for clutch contact problems. Furthermore, the state-of-the-art of the clutch friction characteristics is established that leads to the derivation of a friction model appropriate for clutches, wherein some considerably important friction phenomena are taken into account. Subsequently, the effects of two major friction material degradation mechanisms, namely (i) adhesive wear and (ii) thermal degradation, on the clutch contact and friction characteristics are theoretically and experimentally investigated, which is the second contribution of this thesis. A novel experimental procedure, which is the third contribution, is proposed in order to systematically evaluate how the sliding friction characteristics of a clutch evolve during the lifetime. This way, the evolution of the frictionmodel parameters can be observed and quantified. A knowledge gained from the aforementioned evolution motivates to model and simulate the dynamic responses of a degrading clutch system. Based on the simulation results, some measurable ``features'' representing the progression of the clutch friction material degradation are derived, which is the fourth contribution. A feasibility study on the derived features recommends to develop two clutch monitoring methods: i.e. (a) post-lockup-features-based method and (b) pre-lockup-features-based method. In the first monitoring method, which is the fifth contribution, the torsional natural frequency and the corresponding damping extracted from the post-lockup torsional vibration signals are proposed as relevant features. The second monitoring method, which is the sixth contribution, utilizes the relative velocity signals of interest for extracting the relevant features, i.e. (i) the normalized engagement duration, (ii) normalized Euclidean distance and (iii) Spectral Angle Mapper (SAM) distance. In order to evaluate and verify the developed monitoring methods, accelerated life tests (ALTs) were carried out on some commercial wet clutches using a fully instrumented SAE #2 test setup which is designed by an industrial partner, Dana-Spicer Off Highway Belgium. The experimental results confirm that the developed monitoring methods are feasible to be used for practical applications. As the last contribution, a data-driven health assessment and prognostics method which can be used to assess the condition and to predict the remaining useful life (RUL) of wet friction clutches is developed. This strategy has been successfully evaluated and verified on some commercially available wet clutches. The evaluation and verification show that the developed strategy is prospective for aiding a condition based maintenance for wet friction clutches.nrpages: 284status: publishe