Model-Free Predictive Control Using Polynomial Regressors

Abstract

: Model-free predictive control directly computes the control input from massive input/output datasets and does not use a mathematical model. In contrast, conventional model predictive control relies on mathematical models. Although the underlying principle of model-free predictive control utilizes linear regression vectors comprising input/output data, it can also be applied to control nonlinear systems. In this study, the linear regression vectors are extended to polynomial regression vectors, improving the control performance. Using numerical simulations, we demonstrate the effectiveness of this approach.

Multi‐input multi‐output model‐free predictive control and its application to wastewater treatment

“…Remark 1 A generalization of previous studies regarding only MIMO systems has been described. For m=1 and p=1, the results are identical .…”

“…As discussed by the authors in , PI control can be utilized to improve the datasets used in model‐free predictive control. We prepared datasets containing the samples ( N =300) of inputs/outputs to add a random excitation signal v ∈ R 2 to .…”

“…1. [29,30] regarding only MIMO systems has been described. For m = 1 and p = 1, the results are identical.…”

“…where K p = 0.004 0 0 0.01 , K i = 0.004 0 0 0.04 (32) 3.4. Model-free predictive control As discussed by the authors in [29,30], PI control can be utilized to improve the datasets used in model-free predictive control. We prepared datasets containing the samples (N = 300) of inputs/outputs to .04], respectively, we obtain the control results shown in Fig.…”

“…In Section 2, recent results are generalized to multi‐input multi‐output (MIMO) nonlinear systems. The size of the optimization variables and the computational time needed for model‐free predictive control increase significantly when applied to MIMO systems in comparison with single‐input single‐output systems.…”