DOB Design for Nonminimum-Phase Delay Systems and Its Application in Multivariable MPC Control

Abstract

This brief first presents an improved disturbance observer (IDOB) design method to deal with disturbance observation for the nonminimum-phase delay systems. Then, an IDOB-based compound control scheme is developed to enhance the disturbance rejection performance of multivariable model predictive control (MPC) under various disturbances and uncertainties. In such an IDOB-MPC scheme, the MPC controller acts as a prefilter and generates appropriate control actions such that a desired setpoint tracking response is achieved. The IDOB is used to estimate the various disturbances dynamically and suppress them by feedforward compensation design. Both theory analysis and simulation comparisons have shown good disturbance estimation and rejection performances of the proposed method.Index Terms-Disturbance observer (DOB), improved DOB (IDOB), model predictive control (MPC), multivariable control, nonminimum phase.

Disturbance Observer-based Subspace Predictive Control of a Pressurized Water-type Nuclear Reactor

“…The estimated disturbances are then combined with the control input to mitigate the effect of actual disturbances on the plant output. Due to its simplicity and effectiveness, the DOB has been integrated with MPC in the process industries in the last two decades with appilcations in grinding mill [9], [10], systems with dead-time [11], flight control [12], static var compensator [13], and selective catalytic reduction denitrification system [14]. These integrated schemes provided improved closed loop performances.…”

Design of a Prediction-Accuracy-Enhanced Continuous-Time MPC for Disturbed Systems via a Disturbance Observer

“…In spite of the promising feature and cheerful results on application of MPC to IES, similar to most of the other advanced feedback control methods, MPC usually asymptotically suppresses the disturbances and uncertainties through a manner of feedback regulation in a relatively slow way [11], [13]. Note that integral control action is one of the most practical approaches utilized for early MPC algorithm to counteract the undesirable influences of disturbances and uncertainties, e.g., see dynamic matrix control [2], generalized predictive control [5], and MPC embedded with an integrator [4].…”

“…To this end, the combination of MPC with DOBC has been extensively investigated to enhance the antidisturbance performances of various kinds of systems (see, e.g., [6], [7], [9]- [13], [15]- [20], and references therein). In general, those MPC approaches can be classified into the following two categories.…”

“…The first category mainly focuses on feedforward compensation for the effects of disturbances and uncertainties [6], [9]- [13], [15]. The design philosophy of those MPC methods is quite straightforward.…”

Continuous model predictive control for linear disturbed systems under mismatching condition

“…In spite of the promising feature and cheerful results on application of MPC, similar with most of other advanced feedback control methods, MPC usually asymptotically suppresses the disturbances and uncertainties through a manner of feedback regulation in a relatively slow way [9], [11]. In general, the controller can not directly and promptly react to suppress strong disturbances and compensate severe plant uncertainties [9].…”