A Rolling PID Control Approach and its Applications

TL;DR

This paper introduces a novel rolling PID control method that updates controller parameters dynamically using observable data, addressing the challenges of parameter tuning in nonlinear systems and demonstrating its effectiveness through experiments.

Contribution

It proposes a data-driven rolling PID control approach that simplifies parameter adjustment for nonlinear systems, overcoming limitations of traditional and optimization-based methods.

Findings

Effective parameter updates in each rolling period

Improved control performance demonstrated experimentally

Addresses nonlinear system control challenges

Abstract

The canonical proportional-integral-derivative (PID) control approach has been widely used in industrial application due to their simplicity and ease of use. However, its corresponding controller parameters are hard to be adjusted, especially for nonlinear systems. The optimization-based method provides a general framework to find optimal PID controller parameters; nevertheless, several disadvantages exist, for example, it is nontrivial to select an appropriate sample size and it is necessary to obtain the global optimal solution but the optimization problem is non-convex, making it hard to achieve. To alleviate the aforementioned limitations, a rolling PID control approach is proposed in this study, in which, at each rolling period, the PID controller parameters are updated using observable data, which can be classified to data-driven control method. The effectiveness of the proposed approach has been validated by experiments.