Background: Nowadays the thermal runaway is a serious obstacle in functioning of modern devices and systems because it leads inevitably to their damage. Insufficient study of this phenomenon does not allow to estimate safely predisposition of various accumulators to thermal runaway. There are few works on mathematical modeling of the thermal runaway phenomenon. The works on the phenomenological description of this phenomenon are also not numerous.
Methods: In this work, patents and literary data are analyzed concerning the factors which could cause the development of thermal runaway in various types of batteries. We have a problem of modeling with incompleteness of initial data in the conditions of uncertainty. We offer a new approach to model thermal runaway phenomenon based on fuzzy logic formalism. As it is convenient to display initial and final data in the table form we chose Excel spreadsheets as the tool of modeling.
Results: Examples of thermal runaway probability modelling during the discharge of Ni-CD batteries were reviewed by means of fuzzy logic methods. Our calculations confirmed the conclusion about the influence of battery operation life on the probability of thermal runaway. In our calculations a significant increase in temperature or capacity had a significant influence upon the probability of thermal runaway in the same conditions. Our results also demonstrate that probability of thermal runaway occurrence grows with the growth of charging voltage. It is consistent with experimental data.
Conclusion: Our results are qualitatively and quantitatively coordinated with known experimental data. The summary of available data found in literature and patents related to the thermal runaway phenomenon allows us to conclude that it is feasible to use fuzzy logic methods to model such phenomenon. Im-plementation of the algorithm in Excel spreadsheets is described.