Batteries, Vol. 7, Web pages 85: Impedance Based Temperatures Estimation of Lithium Ion Cells Using Artificial Nerve organs Networks
Batteries doi: 10. 3390/batteries7040085
Authors: Marco Ströbel Julia Pross-Brakhage Mike Kopp Kai Peter Birke
Tracking the cell temp is critical for battery safety and cell stability. It is not achievable to equip every cell with a temperature sensor in large battery systems such as those within electric vehicles. Apart from this, temperature sensors are usually mounted on the cell surface and perform not detect the core temperature, which could mean detecting an offset due in order to the temperature gradient. Several sensorless methods require great computational effort for resolving partial differential equations or require error-prone parameterization. This paper presents a sensorless temperature estimation method with regard to lithium-ion cells using data from electrochemical impedance spectroscopy in combination with artificial neural networks (ANNs). By training a good ANN with data of 28 cells and estimating the cell temperatures of eight more cells of the same cell kind, the neural network (a simple feed forward ANN with just one hidden layer) was able to accomplish an estimation accuracy associated with & amp; Delta; T= 1 K (10 & amp; #8728; C & amp; lt; T& amp; lt; 60 & amplifier; #8728; C) with low computational effort. The temperatures estimations were investigated for various cell types at different states of charge (SoCs) with different superimposed direct currents. Our method is without a doubt easy to use and can be completely automated, while there is no significant offset in monitoring temperature. In add-on, the chance of using the above mentioned approach to calculate additional battery states such as SoC and condition of health (SoH) is definitely discussed.