Adaptive Tuning of Online Feedback Optimization for Process Control Applications

TL;DR

This paper introduces an adaptive tuning method for Online Feedback Optimization controllers that enhances process control performance without needing detailed system models or extra experiments.

Contribution

It proposes a sensitivity-based adaptive tuning approach that automatically adjusts algorithm parameters using only scalar feedback, improving control performance.

Findings

Adaptive tuning improves control performance over manual methods.

The method requires no additional system information or repeated experiments.

Numerical studies validate the effectiveness on gas lift and reactor processes.

Abstract

Online Feedback Optimization leverages properties of optimization algorithms to develop controllers for systems with limited model availability, which is often the case in process control. The interplay between the parameters of the chosen optimization algorithm, as well as lack of direct connection to the characteristics of the underlying process make their tuning challenging. We propose a method for adaptive tuning of Online Feedback Optimization controllers based on scaled projected gradient descent by using sensitivity of the desired objective to the parameters of the algorithm. The proposed adaptive tuning method limits the operator-tunable parameters to scalar values that represent how much the control inputs and the objective can change between iterations without requiring either additional information about the controlled system or repeated experiments. Numerical studies on a gas lift and a continuously-stirred tank reactor processes confirm that our adaptive scheme improves closed-loop performance of Online Feedback optimization compared to standard manual tuning methods.