Evaluation of accuracy for Bernardi equation in estimating heat generation rate for continuous and pulse-discharge protocols in LFP and NMC based Li-ion batteries

Accurate prediction of heat generation in Li-ion batteries during real driving conditions is essential for an efficient thermal management system. In this study, we verify the applicability of a commonly-used heat generation estimator (i.e., Bernardi equation) in Li-ion batteries. The real-world drive cycles comprise of intermittent discharge pulses as opposed to continuous discharge. Therefore, we consider both continuous and pulsedischarge protocols, and compare the heat generation evaluated through Bernardi equation and direct in-situ measurements. It is observed that for continuous discharge, Bernardi equation predicts the heat generation rate with reasonable accuracy. However, the equation substantially overestimates the heat generation under pulse-discharge protocol (realistic scenarios). The heat generation analysis is performed on two leading Li-ion battery chemistries, i.e., LiFePO4 (LFP) and LiNi0.8Mn0.1Co0.1O2 (NMC). Direct measurement shows deviations from the Bernardi equation to be as high as 26% and 49% for LFP and NMC cells, respectively, under high-rate discharge pulses. For the sake of accurate measurement of heat generation, specific heat capacities of Li-ion cells are evaluated with a combined experimental-numerical approach. The heat generation is examined at different cell temperatures and depth of discharge (DoD) levels, and their individual effects on heat generation are analyzed.