Remaining Useful Life Prediction of Lithium Ion Battery Using Tree based Pipe-Line Optimization Tool

M Chakradhar Reddy,M Nitin Viswanath,P Srinivasa Varma

Accurate estimation of the remaining usable life of a component is critical for the Battery management System. This gives details to operators as to when the part will be replaced. A lot of work has been conducted in recent years on the durability and performance of the Lithium-ion batteries, in particular the remaining useful life estimation. Here, the tree-based pipeline optimization tool (TPOT) is used to estimate the remaining useful life of the batteries. With TPOT, machine learning models are presented as expression trees and optimum pipelines are discovered using an algorithmic search approach called genetic programming. Experimental data simulation is conducted utilizing the NASA Ames Prognostics Center of Excellence Battery Datasets to test the proposed process. Results reveal that TPOT parameter optimization is better than ABC-SVR algorithm optimization. Therefore, the suggested approach achieves strong statistical precision and reliability when used to predict the RUL of LIBs.

Volume 12 | Issue 2

Pages: 1955-1960

DOI: 10.5373/JARDCS/V12I2/S20201240