Multiple Model Adaptive Control Applied to Fractional Order Systems

A new structure to design an adaptive controller for a fractional order system is presented in this paper. The system has two layers, in which the first one consists of multiple models, while the second one interconnects them to provide weighted state estimates. The key challenge is to analyze the stability of multilayer fractional order adaptive processes. The primary objective is to determine the feedback for the computation of the second layer weights. The models from the first layer are used to define the second level adaptive law. The proposed multilayer structure significantly improves the transient characteristics of the fractional order adaptation process, which results in a more effective control system.