A Multi-Output Convolved Gaussian Process Model for Capacity Estimation of Electric Vehicle Li-ion Battery Cells
In electric vehicles (EVs) the battery capacity is a key parameter that must be accurately estimated through the service time of the battery. This paper proposes a new machine-learning model namely a Multi-output Convolved Gaussian-Process (MCGP) model for capacity estimation of lithium-ion (Li-ion) battery cells used in an EV application. The proposed technique can be utilized in enhancing the state-of-charge (SOC) estimation accuracy and moreover it can provide an accurate prediction tool for the remaining useful life (RUL) of a battery cell. The performance of the proposed model is validated using experimental data obtained by cycling two 3.6-V/16.5-Ah Li-ion battery cells. Results show the effectiveness of the proposed model.
Paper
Full text
A Multi-Output Convolved Gaussian Process Model for Capacity Estimation of Electric Vehicle Li-ion Battery Cells
Semantic Scholar · Engineering · 2019
Abstract
In electric vehicles (EVs) the battery capacity is a key parameter that must be accurately estimated through the service time of the battery. This paper proposes a new machine-learning model namely a Multi-output Convolved Gaussian-Process (MCGP) model for capacity estimation of lithium-ion (Li-ion) battery cells used in an EV application. The proposed technique can be utilized in enhancing the state-of-charge (SOC) estimation accuracy and moreover it can provide an accurate prediction tool for the remaining useful life (RUL) of a battery cell. The performance of the proposed model is validated using experimental data obtained by cycling two 3.6-V/16.5-Ah Li-ion battery cells. Results show the effectiveness of the proposed model.