Author ORCID Identifier

https://orcid.org/0000-0002-4623-2646

Semester

Spring

Date of Graduation

2023

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

V'yacheslav Akkerman

Committee Co-Chair

Mario Perhinschi

Committee Member

Hailin Li

Committee Member

David Mebane

Committee Member

Omotayo Oshiga

Abstract

Ever-increasing explosions occurring globally at a rapid rate and in diverse situations have re-established the fact that novel, faster, and more accurate approaches must be developed to analyze and possibly curb these explosions and avert their future occurrences. Experimental endeavors and computational fluid dynamics (CFD) simulations as compared to engineering models generally require enormous time, and resources, as well as a high-level of expertise and technicalities. This might, however, delay prompt analysis and the ability to draw conclusions, thereby causing setbacks in recommending safety measures for different situations and conditions. Therefore, robust models which are accurate and fast to give an insight into an explosion, and possibly simulate various conditions easily and quickly, are critically needed. Also, this dissertation develops a comprehensive approach towards explosions, by not only considering characterizing the contemporary gaseous explosions in an enclosure at various configurations but extending the approach to prediction and characterization of the explosions of lithium-ion batteries (LIBs). As we move (paradigm shift) towards the renewable energy age, fire explosions from energy storage systems such as LIBs are on the rise. This makes computational tools a timely, robust, and versatile mechanism for predicting LIBs explosions, characterizing them quickly, and accurately, thereby giving an edge in prompt LIB explosions analysis and helping to proffer potential solutions, as well as aiding future design considerations and safety analyses. However, there are limited works and numerical models that have attempted to quantify and characterize the hazards associated with the explosion of gases ejected from LIBs during thermal runaway inside the battery pack enclosure. Therefore, in this dissertation, LIBs explosion hazards are analyzed, and a computational model is developed to study the explosion venting scenarios of hazardous gases released from LIBs during failure, as well as developing sub-models to characterize the LIB failure, explosions, and their associated hazards. The model developed showed good accuracy and present itself as an easy-to-use, fast, and reliable tool to predict explosion characteristics of both single-compound and multi-compound fuel-air mixtures including LIBs vented gases. The model was further enhanced by integrating a machine learning model capable of predicting laminar flame speed (a key parameter in explosion modelling) of both single and multiple-compound fuel-air mixtures. The effect of various parameters such as vent size, battery chemistry, CO2 concentration, and the state of charge on LIB explosions National Fire Protection Association (NFPA) reduced pressure was also scrutinized.

The results of this dissertation can be integrated into LIBs battery management systems algorithm, to develop safer and more robust models for averting battery failures as well as mitigating the severity of possible explosions. Furthermore, this work will aid LIB safety analysis and can also be employed in the design of safety vents used in LIBs energy storage compartments to mitigate the effects of explosions. For practical scenarios, the NFPA standards are widely used in industry because of their simplified equations. Therefore, the numerical model developed in this dissertation will enable a field engineer, who does not have experience in numerical methods or/ and the physics of complex explosions to use this code to design a vent for a compartment, where an explosion from a LIB failure might occur. Overall, this dissertation will greatly support the fire safety research community in terms of LIB safety analysis and explosion characterization.

Download

Share

COinS