Real time estimation of the wheel-rail contact condtions using multi-Kalman filtering and fuzzy logic

This paper presents a novel technique for the real time estimation of the contact conditions by using a combination of multi-Kalman filtering and fuzzy logic approach. The proposed solution exploits the variations in the dynamic behaviour of a railway wheelset with the changes in wheel-rail contact condition. The proposed system involves the use of multiple model based estimation of the wheelset dynamics in response to different track conditions. Each of the estimators is tuned to match one particular track condition to give the best results at the specific design point. Residuals of each filter are calculated and the level of matches/mismatches is reflected in the residual values of the models concerned. The residuals from all the models are then be assessed by a fuzzy inference system to determine the present operating condition and hence to give real time information about the track conditions.