Direct adaptive constrained receding horizon predictive control with conditional updating application to motor drives with variable inertia

In the case of important parameters variations, the maintenance of a high level of performances may be impossible to reach with a fixed controller, even if the predictive laws ensure intrinsic robustness. A solution is to develop an adaptive version of the predictive controller for systems with parametric disturbances. From the initial work of Wang and Henriksen (1992) in the generalised predictive control case, this paper presents a direct version of adaptive constrained receding horizon predictive control (DACRHPC). The algorithm is rewritten in an original form minimising a performance index, including controller parameters online identification with conditional updating, taking into account terminal constraints. An application to motor drives with flexible transmission is developed, considering some variations of inertia. It is shown that the DACRHPC algorithm improves the performance, compared to the fixed CRHPC as well as classical direct adaptive GPC strategies