Lithium – ion batteries provide best service to the users, more economically and efficiently since they are designed and manufactured imbibing latest technology. It is important to describe the risk as the effect of uncertainty on objectives and current practices in safety engineering which are built from a foundation of managing since uncertainties.

Risk management seeks to:

Risk Analysis in Lithium Ion Battery Packs

One foundational principal of current risk analysis is a focus on observable quantities (e.g. failure occurrence rate) that describe the states of the system. Such observable quantities can be predicted through design and historical data analysis, with the related mathematical uncertainties expressed as probabilities. One specific risk management and analysis tool is Probelolistic Risk Assessment (PRA). It is also called Quantitative Risk Assessment.

Use of PRA

It is commonly used in safety engineering across domains (e.g. Aviation) PRA attempts to capture and mathematically express the current state of knowledge about system including uncertainties. If identifies hazards, their deterministic causes and consequences and provides a way of describing uncertainty. PRA enables the calculation of expected risk values (defined as probability of an event multiplied by the severity of its consequences) so that alternatives can be compared on a similar numerical basis, where there is insufficient data to directly predict behavior. Therefore risk, PRA relies on Fault Tree Analysis (FTA) and Event Tree Analysis (ETA). To construct a system into components this can be more readily quantified.

Total risk is calculated through mathematical functions of the system’s architecture and risk is identified at the component level. PRA logic suggests that for safety engineering risk reduction is equated with improved safety. Despite a history of successful and useful application in and across safety domains, PRA encounters several problems when applied to safety of complex system. These problems stem from the structural assumptions and underlying biases inherent in PRA logic.

PRA Performance

Performing PRA assumes that there is sufficient input/output data and knowledge of the underlying mechanism to make accurate predictions of system behaviour. But there exist many factors effecting safety that are difficult and are further impossible to measure, quantify and therefore observe. These factors challenge that assumptions of PRA and call in to questions to accuracy, the safety predictions proposed for Lithium Ion Battery Packs.

Minute manufacturing variations, untracked environmental conditions (including during shipping and installation) imperceptible chemical side reactions, digital errors whose records and effects are erased biological sensory perception, human understanding and organizational safety culture are all factors known to affect safety in ways that are difficult or impossible to directly observe and predict. Software performance in automated system can also be difficult to observe on in a safety context due to the complexity of system requirements and interactions across organizations involved in its development.

When system risk is calculated based on system architecture and component risk values, risk management assumes system interactions can be combined in deterministic or predictable ways. But today’s complex system increasingly includes social and organizational influences.

Learning Energy Sufficiency of Lithium Ion Battery Packs

To explore whether Lithium-Ion battery packs possess sufficiently observable risk and/or predictably compounded risk amendable to PRA, two examples can be revisited in the context of PRA. These examples come from the aviation industry, automobile industry, watches and more on account of the rich data available in this field; however similar cases exist for the use of PRA in grid energy storage.

First FAA experimentation on fire suppression showed how a fire in shipment of Lithium-Ion batteries could be suppressed using oxygen starvation but that doing so could product the conditions for a combustible gas explosion. The calculated PRA probability of this accident scenario is the number of Lithium-Ion batteries that catch fire during the shipment divided by the total number of batteries shipped. The mean result of this equation predicts approximately OR=41 accidents in the 10 years period from 2012 prior to mitigation interventions. Severity of accidents were negligible and was calculated by taking the numerical sum of the estimated costs associated with crew injuries, damage caused (aeroplane and cargo damage + collateral damage etc)

Calculation for OR

OR= BAR x B in Miles Here OR is occurrence rate (accidents per 10 year period 2012 to 2022) BAR is Battery Accident Rate B ion Miles is Battery Ion Miles

newadmin