Abstract
This paper presents a weighted multiple model adaptive control (WMMAC) scheme to deal with large parametric uncertainty of continuous-time plant. In this proposed scheme, each `local' controller is designed according to the mixed-μ-synthesis method to consider small uncertainty of plant parameters and disturbance; the weighting algorithm is directly based on model output errors rather than the residuals generated by multiple Kalman filters as in classical multiple model adaptive control (CMMAC). The closed-loop stability (signal boundedness) and tracking performance of the proposed WMMAC system are proved with the help of virtual equivalent system (VES) concept and methodology.
Highlights
Many efforts have been made towards ‘‘robust’’ and ‘‘adaptive’’ characteristics in control field
RMMAC differs from classical multiple model adaptive control (CMMAC) mainly in the following three aspects: 1) As local controller strategy, output dynamic compensator designed by mixed-μ-synthesis method replaced LQ state feedback controller in CMMAC; 2) RMMAC separates control from Kalman filters, which are used for state estimation and model identification; 3) Performance-driven methodology to determine the number of required models
MAIN RESULTS we present the stability and tracking performance analysis of the weighted multiple model adaptive control (WMMAC) system based on virtual equivalent system (VES) concept and methodology
Summary
Many efforts have been made towards ‘‘robust’’ and ‘‘adaptive’’ characteristics in control field. AMC approach was proposed with stability and robustness analysis, but the required conditions on the plant, the estimator, and the mixer, are still to be relaxed to make it applicable to more practical situations. As a new concept of adaptive control, reference [10] proposed a novel WMMAC approach that makes weighted sum directly on the multiple model parameter vectors as plant parameters estimates. Compared to classical adaptive control schemes, this approach has faster estimates convergence rate, resulting in better closed-loop performance. This approach requires that the parameter vector of the plant lies in a convex hull of the parameter vectors of multiple models.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
