Abstract
Measured disturbances are often included in model predictive control (MPC) formulations to obtain better predictions of the future behavior of the controlled system, and thus improve the control performance. In the prediction model, a measured disturbance is in many ways treated like a control input to the system. However, while control inputs change only once per sampling interval as new control inputs are calculated, measured disturbances are typically sampled from continuous variables. While this difference is usually neglected, it is shown in this paper that taking this difference into account may improve the control performance. This is demonstrated through two simulation studies, including a realistic multivariable control problem from the petroleum industry. The proposed method requires only a minor modification in the implementation of the prediction model, and may thus improve the control performance with a minimal effort.
Highlights
The very foundation of model predictive control (MPC) is to predict the future behavior of a system based on a model [1]
The results in this paper show that a system discretized using ZOH is quite inaccurate when a measured disturbances sampled from a continuous variable is applied as an input
It was shown that discretizing the system using First-Order Hold (FOH) would be more precise, but quite impractical in the MPC framework
Summary
The very foundation of model predictive control (MPC) is to predict the future behavior of a system based on a model [1]. In order to improve the control performance, feedforward from measured disturbances may be included in this prediction model. This requires that the prediction model includes the dynamics from the measured disturbance to the output, in addition to the dynamics from the control input to the output. Predictions of the output response from measured disturbances may be made in the same manner as with the inputs [2, 3]. There are two fundamental dierences between control inputs and measured disturbances in the MPC framework: 1. While future control inputs are decision variables in the MPC formulation, and are known (predicted), future measured disturbances are unknown to the controller. Control inputs typically change only once per sampling interval (as a new input is calculated and applied), while disturbances are typically sampled from variables that change continuously between samples
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.
