Reliability and Degradation Modelling for Rechargeable Battery

Description

The ubiquitous use of battery-powered electronic devices has created a strong demandof sophisticated battery management systems (BMSs) to maintain battery safety andreliability. Prognostic and health management (PHM), a framework offeringcomprehensive yet individualized solutions for managing system health, has beensuccessfully applied in BMSs. Nevertheless, the increasingly complex battery systemspose significant barriers to existing PHM methods for battery status evaluation, due tothe fact that these methods are often empirical and population based. As a consequence,the estimation and prediction of battery status might be highly biased and thus lead tosafety hazard and other problems in practical operations.Motivated by the new challenges encountered in existing BMSs, the proposed researchdevelops a radically new approach for monitoring and evaluating battery health statusby incorporating the advances in PHM modeling and analysis techniques. The proposedresearch approach focuses on effective and efficient estimation of battery health statusbased on integration of empirical knowledge and real-time data, and heterogeneousinformation on each individual, including the actual field operating conditions andambient environment of a battery system. It is a worldwide trend in PHM research toshift the traditional paradigm from empirical to data fusion and from population basedto individual based. The proposed research is promising across a wide range of battery-poweredapplications.The health status of rechargeable battery is mainly characterized by two keyparameters: state of health (SOH) and state of charge (SOC). SOH denotes theremaining performance of a battery over its whole life cycle, which is usually quantifiedby remaining useful life (RUL), while SOC quantifies the remaining usable energy at thepresent cycle. The main objective of this research is to develop innovative modelingmethods for estimating SOH and SOC by incorporating additional information and PHMadvances into the modeling process. In particular, we will (i) develop degradation modelsand estimation methods for reliable RUL prediction of rechargeable batteries; (ii)develop effective and robust methods that provide accurate estimation for SOC ofrechargeable batteries, particularly under a dynamic ambient environment; (iii) evaluateand test the proposed methods using both experimental and field data, and comparetheir performance with the state of art methods.?