Use this URL to cite or link to this record in EThOS:
Title: Adaptive techniques for estimation and online monitoring of battery energy storage devices
Author: Nejad, S.
ISNI:       0000 0004 6421 8535
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
The battery management system (BMS) plays a defining role in the safety and proper operation of any battery energy storage system (BESS). Without significant advances in the state-of-the-art of BMS algorithms, the future uptake of high power/energy density battery chemistries by consumers in safety-critical applications, is not feasible. Therefore, this thesis aims to provide a coherent body of work on the enhancement of the most important tasks performed by a modern BMS, that is, the estimation and monitoring of various battery states, e.g. state-of-charge (SOC), state-of-health (SOH) and state-of-power (SOP). The Kalman Filter is an elegant set of robust equations that is often utilised by designers in modern BMS, to estimate the battery states and parameters in real time. A nonlinear version of the KF technique, namely the Extended Kalman Filter (EKF) is applied throughout this thesis to estimate the battery’s states including SOC, as well as the battery’s impedance parameters. To this end, a suitable model structure for online battery modelling and identification is selected through a comparative study of the most popular electrical equivalent-circuit battery models for real-time applications. Then, a novel improvement to the EKF-based battery parameters identification technique is made through a deterministic initialisation of the battery model parameters through a broadband system identification technique, namely the pseudorandom binary sequences (PRBS). In addition, a novel decentralised framework for the enhancement of the EKF-based SOC estimation for those lithium-ion batteries with an inherently flat open-circuit voltage (OCV) response is formulated. By combining these techniques, it is possible to develop a more reliable battery states monitoring system, which can achieve estimation errors of less than 1%. Finally, the proposed BMS algorithms in this thesis are embedded on a low-cost microprocessor hardware platform to demonstrated the usefulness of the developed EKF-based battery states estimator in a practical setting. This a significant achievement when compared to those costly BMS development platforms, such as those based on FPGAs (field-programmable gate arrays).
Supervisor: Gladwin, D. T. ; Stone, D. A. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available