Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.04800 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908437539979264 |
|---|---|
| author | Tanjavooru, Vivek Teja Pant, Prashant Hamacher, Thomas Hesse, Holger |
| author_facet | Tanjavooru, Vivek Teja Pant, Prashant Hamacher, Thomas Hesse, Holger |
| contents | This paper presents a mixed-integer, nonlinear, multi-objective optimization strategy for optimal power allocation among parallel strings in Battery Energy Storage Systems (BESS). High-fidelity control is achieved by co-simulating the optimizer with a BESS electro-thermal simulation that models spatial thermal dynamics of the battery, providing real-time State of Charge (SOC) and temperature feedback. The optimizer prioritizes reliability by enforcing power availability as a hard constraint and penalizing battery thermal derating. Within these bounds, the controller performs a Pareto sweep on the relative weights of inverter and battery losses to balance the trade-off between inverter efficiency and battery efficiency. The inverter loss model is based on an empirical lookup table (LUT) derived from a commercial inverter system, while the battery thermal loss model uses SOC and temperature-dependent internal resistance, with electric current computed from the battery Equivalent Circuit Model (ECM). When the optimization was applied to a two-string BESS, the competing effects of inverter and battery losses on system availability and thermal derating were observed. The balanced operation yielded improvements of 1% in battery efficiency, 1.5% in inverter efficiency, and 2% in derating efficiency, while maintaining higher availability. Additionally, a 5 degrees C reduction in BESS peak temperature also suggests reduced thermal stress without compromising availability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_04800 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-Objective Nonlinear Power Split Control For BESS With Real-Time Simulation Feedback Tanjavooru, Vivek Teja Pant, Prashant Hamacher, Thomas Hesse, Holger Systems and Control This paper presents a mixed-integer, nonlinear, multi-objective optimization strategy for optimal power allocation among parallel strings in Battery Energy Storage Systems (BESS). High-fidelity control is achieved by co-simulating the optimizer with a BESS electro-thermal simulation that models spatial thermal dynamics of the battery, providing real-time State of Charge (SOC) and temperature feedback. The optimizer prioritizes reliability by enforcing power availability as a hard constraint and penalizing battery thermal derating. Within these bounds, the controller performs a Pareto sweep on the relative weights of inverter and battery losses to balance the trade-off between inverter efficiency and battery efficiency. The inverter loss model is based on an empirical lookup table (LUT) derived from a commercial inverter system, while the battery thermal loss model uses SOC and temperature-dependent internal resistance, with electric current computed from the battery Equivalent Circuit Model (ECM). When the optimization was applied to a two-string BESS, the competing effects of inverter and battery losses on system availability and thermal derating were observed. The balanced operation yielded improvements of 1% in battery efficiency, 1.5% in inverter efficiency, and 2% in derating efficiency, while maintaining higher availability. Additionally, a 5 degrees C reduction in BESS peak temperature also suggests reduced thermal stress without compromising availability. |
| title | Multi-Objective Nonlinear Power Split Control For BESS With Real-Time Simulation Feedback |
| topic | Systems and Control |
| url | https://arxiv.org/abs/2507.04800 |