Sample-Efficient Counterfactual Tuning for Compressor Pressure Control

TL;DR

This paper presents a data-driven, counterfactual tuning method for compressor pressure control that efficiently retunes controllers using historical data, ensuring safety and stability without explicit plant models.

Contribution

It introduces a novel sample-efficient retuning approach based on counterfactual explainability, avoiding the need for explicit plant models or prior control laws.

Findings

Effective in achieving desired performance with minimal controller adjustments

Ensures stability during retuning without costly plant disruptions

Validated through extensive Monte Carlo simulations

Abstract

In controlled industrial environments, ensuring safety and performance during controller tuning is a challenging and critical task. In particular, control loops in compressor-plenum-throttle systems cannot tolerate costly interruptions, and aggressive excitation may lead to unsafe operating regimes. Given the wide availability of historical data, this paper introduces a counterfactual explainability approach for sample-efficient retuning of compressor control loops. The proposed data-driven algorithm determines, without an explicit plant model or previous control law, the smallest controller adjustment required to achieve predefined performance specifications while guaranteeing stability. The effectiveness of the method is demonstrated through an extensive Monte Carlo simulation study.