Remaining Lifetime Prediction for Reliability-based Dry Clutch Control



Published Jul 4, 2018
Daniel Strommenger


Increasing requirements for reliability of modern powertrains
can be achieved by predictive maintenance and reliabilitybased
control based on lifetime prediction. This contribution
presents lifetime prediction for a dry clutch, being an essential
component of automated manual transmissions. Modelbased
development of lifetime prediction requires knowledge
of dry clutch wear, which was identified in previous experiments.
The derived wear model allows estimation of characteristic
wear-dependent values, like friction lining material
losses and friction coefficient changes. Based on these estimated
values the presented lifetime prediction was developed
by fusing these estimated values into a health index (HI) describing
the systems healthiness. Furthermore, the remaining
useful lifetime (RUL) becomes predictable from observations
of health index trend using an exponential weighted
moving average. Eventually, the presented lifetime prediction
was implemented and tested on real-time operating hardware
similar to common transmission control units. In order to
control the system lifetime in normal operation, target trends
for health index and predicted remaining useful lifetime were
defined. Based on trend deviations, a fuzzy-logic based control
strategy was realized, which sets the optimization target
for a reliability-based control. Thus, the optimization target
can be varied between comfort-optimized or wear-optimized
clutch engagements. Finally, an outline of reliability-based
control concepts is given.

How to Cite

Strommenger, D. (2018). Remaining Lifetime Prediction for Reliability-based Dry Clutch Control. PHM Society European Conference, 4(1).
Abstract 392 | PDF Downloads 1606



Remaining Lifetime Prediction, Reliability-based Control, Health Assessment, Wear Modeling, Dry Clutch

Technical Papers