Optimal Tracking Current Control of Switched Reluctance Motor Drives Using Reinforcement Q-learning Scheduling

TL;DR

This paper introduces a novel Q-learning based scheduling method for current control in switched reluctance motor drives, enabling model-free, optimal tracking of reference currents with smooth transitions, validated through simulations and experiments.

Contribution

It proposes a new scheduled-Q-learning algorithm that operates without model parameters to effectively track current trajectories in SRMs.

Findings

Successful simulation validation of the control scheme

Experimental results confirm improved current tracking performance

Smooth transition between control states achieved

Abstract

In this paper, a novel Q-learning scheduling method for the current controller of switched reluctance motor (SRM) drive is investigated. Q-learning algorithm is a class of reinforcement learning approaches that can find the best forward-in-time solution of a linear control problem. This paper will introduce a new scheduled-Q-learning algorithm that utilizes a table of Q-cores that lies on the nonlinear surface of a SRM model without involving any information about the model parameters to track the reference current trajectory by scheduling infinite horizon linear quadratic trackers (LQT) handled by Q-learning algorithms. Additionally, a linear interpolation algorithm is proposed to guide the transition of the LQT between trained Q-cores to ensure a smooth response as state variables evolve on the nonlinear surface of the model. Lastly, simulation and experimental results are provided to validate the effectiveness of the proposed control scheme.