Clutch wear prediction via a deterministic-stochastic modeling approach

The front surface temperature of a clutch pressure plate is studied for clutch wear prediction. A combined deterministic plus stochastic modeling approach is proposed to fit the front surface temperature data. The underlying clutch engagement process is explained by a deterministic function, whereas the inherent dynamics are explained by the stochastic model. The results of the combined modeling approach indicate that the front surface temperature can be modeled accurately to within ± 4°F.