Robust Neural Policy Distillation of Long-Horizon FCS-MPC for Flying-Capacitor Three-Level Boost Converters

TL;DR

This paper presents a neural network-based approach to emulate long-horizon FCS-MPC for flying-capacitor converters, enhancing robustness and reducing computational costs while maintaining stability under various conditions.

Contribution

It introduces a robust neural policy distillation method using DAgger for flying-capacitor converters, enabling efficient and stable control with transferability to similar systems.

Findings

Neural policy maintains stable voltage regulation and capacitor balancing.

The approach reduces computational burden compared to traditional methods.

Transfer learning improves sample efficiency in related converter systems.

Abstract

Long-horizon finite-control-set model predictive control (FCS-MPC) can improve transient regulation and flying-capacitor balancing in flying-capacitor three-level boost converters (FC-TLBCs). However, searching over switching sequences becomes computationally expensive at high switching frequencies. We train a feedforward neural network to imitate an $N$-step FCS-MPC expert computed with beam search. To improve robustness, expert trajectories are generated under randomized input voltage, load resistance, and component parameters, and a disagreement-based DAgger variant is used to relabel on-policy states where the student and expert disagree. In simulation, the learned policy maintains stable voltage regulation and capacitor balancing under nominal conditions, operating-point changes, and perturbations of several physical parameters. We demonstrate the effectiveness of our approach by reducing the computational burden. We also demonstrate transfer to an NPC-type three-level buck converter, where initializing from the FC-TLBC network improves sample efficiency compared with training from scratch.