Fig. 2: Temperature estimation accuracy comparison between LP, FNN, and LP+FNN models. | Communications Engineering

Fig. 2: Temperature estimation accuracy comparison between LP, FNN, and LP+FNN models.

From: Thermal fault detection of lithium-ion battery packs through an integrated physics and deep neural network based model

Fig. 2

a, b Performance comparison for the test profile with US06 discharge and 6C charge profiles at 15 °C on cell#1. a, The US06 discharge and 6C charge profile. b Temperature estimation accuracy comparison on cell#1. c Temperature estimation RMS errors for all testing profiles on 8 cells. The average RMSE for LP, FNN, and LP+FNN models are 1.30 °C, 1.84 °C, and 0.45 °C, respectively.

Back to article page