Optimal PID and Antiwindup Control Design as a Reinforcement Learning Problem
This work addresses interpretability and stability issues in DRL-based control for industrial applications, though it appears incremental by adapting existing frameworks to simpler controller structures.
The paper tackled the problem of interpretability and stability in deep reinforcement learning (DRL) for process control by modeling linear fixed-structure controllers like PID as shallow neural networks, resulting in a method that is inherently stabilizing during and after training and amenable to known operational gains.
Deep reinforcement learning (DRL) has seen several successful applications to process control. Common methods rely on a deep neural network structure to model the controller or process. With increasingly complicated control structures, the closed-loop stability of such methods becomes less clear. In this work, we focus on the interpretability of DRL control methods. In particular, we view linear fixed-structure controllers as shallow neural networks embedded in the actor-critic framework. PID controllers guide our development due to their simplicity and acceptance in industrial practice. We then consider input saturation, leading to a simple nonlinear control structure. In order to effectively operate within the actuator limits we then incorporate a tuning parameter for anti-windup compensation. Finally, the simplicity of the controller allows for straightforward initialization. This makes our method inherently stabilizing, both during and after training, and amenable to known operational PID gains.