Modern electric locomotives have to transfer high tractive effort between wheels and rails. Therefore, they have to be equipped with slip controllers that enable to achieve required force and keep wheelsets velocity in the acceptable value. The slip controller is a part of the locomotive control algorithm that runs on an axle computer that is typically a digital signal processor with limited computational power. The modern slip controller requires determining an actual value of adhesion that cannot be directly measured during the train runs. Therefore, the adhesion estimation is necessary. Extended Kalman filter or unscented Kalman filter can be used with an advantage because they can cope with some drawbacks of conventional methods. However, they are not typically used because of their high computational requirements. Therefore, the extended Kalman filter and unscented Kalman filter are implemented into a digital signal processor that is typically used in the locomotives as an axle computer to verify the possibility of used slip control methods based on the mentioned estimators in the paper. The verification is made on measured data.
Keywords-slip control; extended Kalman filter;unscented Kalman filter; railway.Where φD-W is a difference angle in the different ends of the shaft, ωM is the motor angular velocity, ωW is the wheel angular velocity, i is estimated adhesion, cMW and dMW is a stiffness and damming of the shaft between the motor and the wheel, JM and JW are moments of inertia of the motor and wheel respectively and rG is a gear ratio of the gearbox.