Discretization and Digital Cascaded Controller Designs for MIMO Analog Systems with Multiple Input Output and State Delays.

摘要

Multiple-input multiple-output (MIMO) systems with multiple time delays in states, inputs and outputs are frequently encountered in Networked Control Systems (NCSs).;To address the problem of network-induced time delays, this dissertation proposes two approaches for the delay compensation. One is based on discrete-time state-space modeling. In this work, the Chebyshev quadrature formula together with the Lagrange linear interpolation approach is employed in extending the conventional discrete-time state-space modeling to MIMO systems with multiple delays. The cascaded system is formulated as an augmented discrete time system for state-feedforward and state-feedback linear-quadratic regulator (LQR) design. The parameters of the optimal controller and its associated state-feedback are determined by tuning the weighting matrices in the LQR optimality performance criteria. Another approach is based on the digital redesign technique. Here, an analog cascaded controller is first desigmed for the delay-free analog MIMO system; it is then converted to a corresponding digital one with prediction capability, such that the delayed system can demonstrate performance similar to the delay-free system.;Two kinds of motors are utilized to confirm the theory derivations. The squirrel cage induction motor (SCIM) is selected as the test bed for the proposed discretization modeling and the direct digital control technique. The Permanent Magnet Direct Current (PMDC) motor is employed in the simulation and experimental study for the digital redesign technique. Roth systems are implemented through a dSPACE real-time DSP control platform. Experimental results confirmed the simulation results very well, indicating the applicability and correctness of the proposed methodologies.