Redox flow battery time-varying parameter estimation based on high-order sliding mode differentiators

A new insight into vanadium redox flow batteries (VRFB) parameter estima-tion is presented. Driven by the electric vehicles proliferation, a hybrid fast-charging station with grid and a renewable energy connection is particularlyconsidered. In this stationary application, the VRFB is operating as bu...

Descripción completa

Detalles Bibliográficos
Autores: Fornaro, Pedro, Puleston, Thomas Paul|||0000-0003-3251-7533, Puleston, Paul Federico, Serra, Maria|||0000-0002-9885-8093, Costa Castelló, Ramon|||0000-0003-2553-5901, Battaioto, Pedro
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/372352
Acceso en línea:https://hdl.handle.net/2117/372352
https://dx.doi.org/10.1002/er.8319
Access Level:acceso abierto
Palabra clave:Energy storage
Storage batteries
Parameter estimation
Sliding mode differentiator
State of charge
State of health
Vanadium redox flow battery
Energia -- Emmagatzematge
Acumuladors
Àrees temàtiques de la UPC::Informàtica::Automàtica i control
Descripción
Sumario:A new insight into vanadium redox flow batteries (VRFB) parameter estima-tion is presented. Driven by the electric vehicles proliferation, a hybrid fast-charging station with grid and a renewable energy connection is particularlyconsidered. In this stationary application, the VRFB is operating as bufferingmodule. This hybrid topology could contribute to reduce the grid connectioncost of the charging station. However, to make VRFB a viable technology,improvements are needed. Among these, some of the most important are inthe field of the estimation of the battery's State of Charge, State of Health, andinternal parameters. The proposed estimation method is based on a recursiveleast square (RLS) estimation algorithm with forgetting factor, combined witha sliding mode finite-time convergent differentiation algorithm. The latter pro-vides robust exact derivatives of both VRFB's current and voltage with a highdegree of noise rejection, required by the RLS algorithm to perform a preciseestimation. The proposed sliding mode-based estimation setup is completedwith a systematic methodology to guarantee the validity of the on-line esti-mated values, depending on the persistence of excitation of the measured cur-rent and voltage. Finally, the methodology is thoroughly analysed andvalidated by computer simulation.