The development of Battery Management Systems (BMSs) able to determine the State of Charge (SoC) and the State of Health (SoH) of lithium-ion accumulators through a simple and cost-effective procedure is becoming of paramount importance considering their potential impact on two key enabling technologies of the 21st century, i.e., electrical mobility and energy production from renewable sources. In this paper, a novel machine-learning model, based on the Random Forest (RF) and suitable to BMS, is presented. The model is able to estimate SoC and SoH under different operating conditions by exploiting only impedance measurements derived from Electrochemical Impedance Spectroscopy (EIS). In particular, we provide an assessment of the proposed model by investigating its performance in terms of accuracy and reliability.
A Novel Machine Learning Algorithm for State of Health Prediction of Lithium-Ion Batteries
Battaglia, Filippo;Campobello, Giuseppe;Aloisio, Davide;Leonardi, Salvatore Gianluca;Gugliandolo, Giovanni;Sergi, Francesco;Donato, Nicola
2023-01-01
Abstract
The development of Battery Management Systems (BMSs) able to determine the State of Charge (SoC) and the State of Health (SoH) of lithium-ion accumulators through a simple and cost-effective procedure is becoming of paramount importance considering their potential impact on two key enabling technologies of the 21st century, i.e., electrical mobility and energy production from renewable sources. In this paper, a novel machine-learning model, based on the Random Forest (RF) and suitable to BMS, is presented. The model is able to estimate SoC and SoH under different operating conditions by exploiting only impedance measurements derived from Electrochemical Impedance Spectroscopy (EIS). In particular, we provide an assessment of the proposed model by investigating its performance in terms of accuracy and reliability.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
