State of Charge Estimation of LiFePO4 in Various Temperature Scenarios

Mingzhu Wang, Guan Wang, Zhanlong Xiao, Yuedong Sun, Yuejiu ZhengBatteries 2023 · DOI

TL;DR

This paper tackles the challenge of accurately estimating the State of Charge (SOC) of LiFePO4 batteries under varying temperatures, a critical issue for electric vehicles and energy storage. They propose an improved Extended Kalman Filter (EKF) that dynamically adjusts its gain based on the battery's Open Circuit Voltage (OCV) behavior, achieving SOC estimation errors within 3% across different temperatures and even estimating low-temperature capacity with less than 1% error.

What Problem Does This Paper Address?

Accurate State of Charge (SOC) estimation is crucial for the safe and efficient operation of battery management systems (BMS), especially in electric vehicles (EVs). LiFePO4 batteries, popular for their environmental friendliness and cost-effectiveness, present unique challenges for SOC estimation. Their "voltage plateau" phase, where the voltage barely changes with SOC, makes traditional OCV-based estimation methods unreliable. Furthermore, LiFePO4 batteries exhibit poor performance and parameter variations at low temperatures, compounding the estimation difficulty. Existing methods often struggle to maintain accuracy when faced with both fluctuating ambient temperatures and dynamic load conditions. This paper aims to bridge this gap by developing a robust SOC estimation method that accounts for these temperature-dependent variations and the specific characteristics of LiFePO4 batteries.

Methods

The authors introduce a multi-pronged approach:

  • Temperature-Aware Equivalent Circuit Model (ECM): They establish a second-order RC ECM that explicitly incorporates temperature-dependent parameters.
  • Particle Swarm Optimization (PSO) for Parameter Identification: PSO is employed to efficiently identify the numerous parameters of the temperature-dependent ECM using experimental data.
  • Improved Extended Kalman Filter (EKF): The core of their solution is a modified EKF. Key enhancements include:
  • Online OCV Identification: Using a Forgetting Factor Recursive Least Squares (FFRLS) method to estimate the OCV curve in real-time.
  • Adaptive Kalman Gain Adjustment: The Kalman gain is dynamically adjusted based on whether the battery is in its OCV plateau phase or non-plateau phase. This allows the EKF to rely more on ampere-hour integration during plateaus and the battery model during non-plateau regions.
  • Redefined SOC Limits: New upper and lower SOC limits are defined for different temperatures to better reflect the battery's actual usable capacity under varying thermal conditions.
  • Weighted Iterative Least Squares (WILS) for Capacity Estimation: For low-temperature capacity estimation, they leverage the accurate SOC estimates from the improved EKF and apply the WILS method.

Datasets

The experiments were conducted using a LiFePO4 battery from Tianjin Lishen Battery Company. The paper describes the following experimental procedures:

  • Basic Performance Tests: Conducted at various temperatures (45°C, 25°C, 5°C, -15°C) to characterize battery parameters.
  • Variable Temperature OCV Experiments: To obtain OCV-SOC curves across a range of temperatures and SOC levels.
  • Hybrid Pulse Power Characteristic (HPPC) Tests: Used to determine internal resistance at different temperatures.
  • Dynamic Working Condition Experiments (NEDC): Performed under simulated real-world driving conditions with both rising and falling ambient temperatures to evaluate the SOC estimation under variable temperature and dynamic load.
  • Low-Temperature Capacity Tests: Conducted at 0°C on both fresh and aged batteries to validate the capacity estimation method.

Key Insights and Ideas

  • OCV Plateau is the Achilles' Heel: The paper strongly emphasizes that the OCV plateau is the primary hurdle for accurate SOC estimation in LiFePO4 batteries, and their method directly addresses this by adapting the estimation strategy.
  • Temperature is Not Just a Parameter, It's a System: They demonstrate that temperature significantly impacts battery parameters beyond just OCV, necessitating a temperature-aware model for accurate state estimation.
  • Adaptive EKF is Key: The adaptive Kalman gain, switching between OCV-plateau and non-plateau strategies, is a critical innovation that allows the EKF to navigate the challenging LiFePO4 OCV characteristics.
  • Redefining SOC Limits for Real-World Use: The concept of temperature-dependent SOC limits is insightful, as it acknowledges that the "full" capacity of a battery isn't constant across all thermal conditions.
  • Capacity Estimation Benefits from Accurate SOC: By first achieving reliable SOC estimates, even at low temperatures, they can then more accurately estimate battery capacity, which is also temperature-sensitive.

Key Results

  • SOC Estimation Accuracy: The proposed improved EKF algorithm achieved SOC estimation errors within 3% under variable temperature conditions.
  • Low-Temperature Capacity Estimation: The capacity estimation at low temperatures (0°C) using the IRLS algorithm, built upon the accurate SOC estimates, resulted in errors of less than 1% for both fresh and aged batteries.
  • Robustness: The method demonstrated robustness against initial SOC errors, capacity errors, and simulated sensor noise (current and voltage drift) in variable temperature scenarios.
  • OCV Behavior: The study found that while OCV varies with temperature, the differences were negligible across the tested range, allowing them to use a single OCV-SOC curve (at 25°C) for their model, simplifying the process.

Limitations and Future Work

  • Model Complexity: While the second-order RC model is a good compromise, more complex electrochemical models could potentially offer higher accuracy but at the cost of computational burden.
  • Experimental Range: The experiments were conducted within specific temperature ranges. Further validation at more extreme temperatures might be beneficial.
  • Battery Type: The study focuses on LiFePO4 batteries. While the principles might be transferable, specific parameter tuning would be required for other battery chemistries.
  • Future Work: The authors suggest that their method provides a solid foundation for accurate SOC and capacity estimation under challenging thermal conditions, paving the way for more reliable battery management systems in EVs and energy storage applications. The accurate low-temperature capacity estimation is highlighted as a significant step forward.

Summary updated

Abstract

The state estimation of a battery is a significant component of a BMS. Due to the poor temperature performance and voltage plateau phase in LiFePO4 batteries, the difficulty of state estimation is greatly increased. At the same time, the ambient temperature in which the battery operates is changeable, and its parameters will vary with the temperature. Therefore, it is extremely challenging to estimate the state of LiFePO4 batteries under variable temperatures. In an effort to accurately estimate the SOC of LiFePO4 batteries at different and variable temperatures, as well as its capacity at low temperature, the characteristics of LiFePO4 batteries at different temperatures are first tested. In addition, a variable temperature OCV experiment is designed to obtain the OCV of the full SOC range. Then, the ECM considering temperature is established and all parameters are identified by PSO. Finally, an improved EKF algorithm is presented to accurately estimate the SOC of LiFePO4 batteries at different and variable temperatures. Meanwhile, the battery capacity at low temperature is further estimated based on the estimated SOC result. The results show that SOC estimation errors at variable temperature are all within 3%, and the capacity estimation errors at low temperature are all within 1%.
BibTeX
@inproceedings{wang2023state,
  author = {Mingzhu Wang and Guan Wang and Zhanlong Xiao and Yuedong Sun and Yuejiu Zheng},
  title = {{State of Charge Estimation of LiFePO4 in Various Temperature Scenarios}},
  booktitle = {Batteries},
  year = {2023},
  doi = {10.3390/batteries9010043},
}
Figure