The monitoring of temperature during operation of lithium ion batteries is crucial for a stable and safe performance of a battery system. However, a major challenge is to obtain exact internal temperature progression upon charge and discharge processes. Though commonly used resistance temperature sensors can give considerably accurate temperature values, they are limited to the application outside of the cell due to the distortion of their signal through induction effects of the applied cell currents, if installed inside the pouch cell and therefore, they are not suitable for in situ measurements. This poses serious problems especially for big cell stacks, as core temperatures for those cells differ greatly from the outer parts and particularly the surface of the cell encasement. Fiber Bragg Grating (FBG) sensors on the other hand, provide data that are unaffected by induction effects as their operating principle is solely based on the reflection of light of certain wavelengths within a glass fiber. However, a specific difficulty is the distinction between temperature and pressure effects, as the reflected signal is sensitive to both. Often times additional tubing around the sensor is needed to eliminate pressure effects or an additional sensor is placed in close proximity, inside or outside, to the FBG-sensor as a reference. The former, however, can affect the cell stack physically acting as a defect spot with unwanted side effects.

In our work, we show that a careful calibration of the FBG-sensor can be accomplished, so that both effects can be separated and the FBG-sensor can then be used to gather information not only about temperature but also about volume changes in an operating cell. Therefore, a temperature calibration took place, with the cell being held at a constant SoC. Subsequently, a calibration of the SoC - dependent pressure-induced mechanical deformation was performed at very low C-rate to minimize thermal effects. The so gathered calibration parameters can then be applied to all performed cycling tests, regardless of the C - rate. A calibration routine and mathematical solution to separate the influences of temperature and pressure to receive exact in situ temperature values for a 10 Ah lithium ion battery cell is presented. The calibration method is applied to cycling and rate tests at various C-rates.

The results show, that wavelength shifts upon charging and discharging processes are mainly resulting from volume changes of the active cell components, meaning that the FBG – sensor has a very high sensitivity towards pressure changes. With the before mentioned calibration method the impact of mechanical deformation on the FBG - sensor can be subtracted from the raw data and allow for a direct temperature reading. The temperature data correlates with data gathered from commonly used sensors placed on the outside of the pouch cell. Tests with varying discharge rates confirm these results and show, that the method produces precise temperature data even at different C-rates, even though only slow C-rates were applied during calibration.

In detailed comparison, the FBG - sensors show a faster response towards temperature changes compared to conventional external temperature sensors, which can be related to the insulating effect of the pouch foil. In case of high discharge rates, we were able to show that the measured temperatures inside of the pouch cell are higher in comparison to the outside ones. Furthermore, SoC - induced changes in pressure measured by two FBG-sensors, which were located at different positions on the cell stack, are almost identical. This indicates that pressure build-up within a pouch cell occurs evenly through the cell stack.