A Systematic Review of Thermal Runaway in Li-ion Batteries: Pathways, Detection Techniques, and Early Warning Models

In the recent past, various accidents related to lithium battery fires have been reported worldwide, some of them being fatal. This emphasizes the need to improve battery systems' safety for their users. To achieve this, first, an understanding of the phenomena that take place during TR is essential. Then, such an understanding may be used to identify sensors and models that can predict TR, with sufficient anticipation. Following this logic, this systematic literature review thoroughly analyzes the state-of-the-art regarding the different abuses and the consequent safety threats to lithium-based battery cells to (i) describe the various phenomena that occur when a battery cell is abused, (ii) analyze the different sensors that are found in the literature which can be used to detect a faulty or abused battery cell, and (iii) review the various models that are used to analyze the data acquired by the sensors. Since this is a systematic literature review, the methodology followed to carry out this review is rigorously described, including the literature database search string and paper-inclusion criteria used to screen the list of works and arrive at a final set of meaningful papers. Finally, in view of the findings, a combination of sensors and models is suggested to assess the safety level of a battery cell. It is concluded that temperature monitoring, along with an empirical model, is best suited for abuse detection, while voltage monitoring, along with a statistical model, should be adopted for cells' fault detection.