Data-driven health monitoring and control of battery energy storage system in distribution networks
Li-Ion batteries (LIBs) have been widely utilized in electric vehicles and battery energy storage systems (BESS) for power grid applications. To avoid system failure through timely maintenance, it is of great importance to estimate the battery state of health (SOH). This thesis first develops data-d...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Doctor of Philosophy |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/159830 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Li-Ion batteries (LIBs) have been widely utilized in electric vehicles and battery energy storage systems (BESS) for power grid applications. To avoid system failure through timely maintenance, it is of great importance to estimate the battery state of health (SOH). This thesis first develops data-driven methods for online SOH estimation of LIBs under different load profiles. Novel health indicators (HIs) that are extracted from various load profiles are proposed and advanced machine learning algorithms are developed to map the relationship between HIs and SOH. Then, this thesis identifies the coupling relationship between frequency regulation and voltage regulation in low-voltage distribution networks with relatively low x/r ratio. To this end, a fuzzy logic-based controller is proposed to provide coordinated frequency and voltage support via BESS. To mitigate battery health degradation, a battery lifetime model is built and used to design fuzzy rules of the controller. |
|---|