Selected Control Strategies for Nonlinear Technological Processes

Abstract

Nonlinear dynamics are frequently encountered in industrial processes, where conventional linear controllers often fail to ensure the required performance or safety due to phenomena such as multiple equilibria, limit cycles, and bifurcations, all of which can negatively impact system stability. These challenges necessitate control strategies that can adapt to changing operating conditions and capture nonlinear behavior more effectively. Building on earlier research, this study identifies and evaluates two promising approaches to model-based adaptive control suitable for nonlinear technological processes. 

Within an established classification framework, two representative strategies are examined: one based on sequential linearization combined with polynomial control, and another employing a direct nonlinear method based on the Wiener model. Experimental validation is conducted on a two-tank benchmark and a continuous stirred-tank reactor, both representing typical nonlinear industrial systems.

The results show that sequential linearization in combination with polynomial control achieves faster set-point tracking, quicker convergence, and greater robustness compared to Wiener model control. These findings underscore the advantages of structured sequential linearization techniques within adaptive control and demonstrate their potential as a practical and effective alternative to purely linear control designs in complex nonlinear environments.