Multi-Layer Control for Hybrid Balancing Systems

doi:10.1109/CCTA48906.2021.9658882

Multi-Layer Control for Hybrid Balancing Systems

2021-08-09

2

PRE

AI

Jorge Varela Barreras

H

Herschel C. Pangborn

DOI：10.1109/CCTA48906.2021.9658882

Original

Original request for help

Save

Abstract

En 中文

Hybrid balancing is a recently-proposed class of battery balancing systems that simultaneously provide capacity and thermal equalization, while enabling hybridization with supercapacitors. This integration of functions poses a challenging control problem, requiring the fulfillment of multiple objectives (e.g., reduction of charge and temperature imbalances, energy losses and battery stress) and the coordination of a large number of power converters. To tackle this challenge, we propose a multi-layer model predictive control (MPC) framework, which splits the control tasks into two layers. The first layer uses long prediction horizons and a simplified model of the energy storage system to compute the state-of-charge reference for the supercapacitors. The second layer uses module-level models of the battery pack to track this reference, while minimizing charge and temperature imbalances with a small prediction horizon. Simulation results demonstrate that the multi-layer MPC provides similar performance as single-layer MPC, but at a fraction of the computational effort.

Keywords:

MANAGEMENT

CHARGE

Multi-Layer Control for Hybrid Balancing Systems

Abstract

Journal

Organization