Application of direct adaptive generalized predictive control to an automatic gear box with a continuous variable transmission

Even if predictive control has proved to be an efficient strategy, the maintenance of a high level of performances may be impossible to reach with a fixed predictive controller in case of important parameter variations. A solution is then to develop an adaptive version of the predictive controller for systems with parametric disturbances. This paper presents a direct version of adaptive generalized predictive control, based on the work of Wang and Henriksen. The algorithm is rewritten in an original form minimizing a performance index, using a least-squares type strategy for the controller parameter online identification, including a conditional updating test in the adaptation loop. An application of this structure to an engine drive with an automatic gear box with continuous variations of the reduction ratio is finally developed. The model of this system is derived as a starting point from the well-known two-masses problem. In this study, only the engine speed is measured, the acceleration (variable to be controlled) remains not measurable.