Machine Learning based prediction of Vanadium Redox Flow Battery temperature rise under different charge-discharge conditions

TL;DR

This study demonstrates the use of machine learning algorithms to accurately predict the temperature rise in Vanadium Redox Flow Batteries during charge-discharge cycles, aiding thermal management.

Contribution

First application of ML algorithms to predict VRFB temperature behavior during operation, comparing LR, SVR, and XGBoost for the first time.

Findings

XGBoost achieved around 99% prediction accuracy.

ML predictions can help optimize VRFB thermal management.

Support Vector Regression and Linear Regression showed lower accuracy.

Abstract

Accurate prediction of battery temperature rise is very essential for designing an efficient thermal management scheme. In this paper, machine learning (ML) based prediction of Vanadium Redox Flow Battery (VRFB) thermal behavior during charge-discharge operation has been demonstrated for the first time. Considering different currents with a specified electrolyte flow rate, the temperature of a kW scale VRFB system is studied through experiments. Three different ML algorithms; Linear Regression (LR), Support Vector Regression (SVR) and Extreme Gradient Boost (XGBoost) have been used for the prediction work. The training and validation of ML algorithms have been done by the practical dataset of a 1kW 6kWh VRFB storage under 40A, 45A, 50A and 60A charge-discharge currents and 10 L min-1 of flow rate. A comparative analysis among the ML algorithms is done in terms of performance metrics such as correlation coefficient (R2), mean absolute error (MAE) and root mean square error (RMSE). It is observed that XGBoost shows the highest accuracy in prediction of around 99%. The ML based prediction results obtained in this work can be very useful for controlling the VRFB temperature rise during operation and act as indicator for further development of an optimized thermal management system.