VABO: Violation-Aware Bayesian Optimization for Closed-Loop Control Performance Optimization with Unmodeled Constraints

Abstract

We study the problem of performance optimization of closed-loop control systems with unmodeled dynamics. Bayesian optimization (BO) has been demonstrated effective for improving closed-loop performance by automatically tuning controller gains or reference setpoints in a model-free manner. However, BO methods have rarely been tested on dynamical systems with unmodeled constraints. In this paper, we propose a violation-aware BO algorithm (VABO) that optimizes closed-loop performance while simultaneously learning constraint-feasible solutions. Unlike classical constrained BO methods which allow an unlimited constraint violations, or 'safe' BO algorithms that are conservative and try to operate with near-zero violations, we allow budgeted constraint violations to improve constraint learning and accelerate optimization. We demonstrate the effectiveness of our proposed VABO method for energy minimization of industrial vapor compression systems.

Constrained Efficient Global Optimization of Expensive Black-box Functions

“…In black-box optimization problems, we typically need to sequentially evaluate different candidate solutions to find the optimal one without explicit gradient information. For example, hyperparameter tuning in machine learning models [1,2], control system performance optimization [3,4] and drugs discovery or materials design [5,6] can all be formulated as black-box optimization problems.…”

“…A key challenge in many applications is black-box constraints [4]. For example, when tuning the temperature controller parameters of a building, we need to minimize the energy consumption while keeping the comfort at user-defined levels.…”

“…More recently, a third group of approaches have been studied (e.g., [4,17]), where constraint violations are allowed, but limits are placed on the total viola-tion incurred during the optimization process. In this setting, two more recent works adopt a penalty-function approach [18] and primal-dual approach [17] respectively to solve constrained efficient global optimization problems.…”