A New Health Analysis Method for Lithium-Ion Batteries Based on the Evidential Reasoning Rule Considering Perturbation

Yunyi Zhang, Shuchang Wang, Wei He, Wei Zhang, Shuaiwen Tang, Guohui ZhouBatteries 2023 · DOI

TL;DR

This paper introduces a novel health assessment method for lithium-ion batteries that leverages the Evidential Reasoning (ER) rule to integrate key performance indicators and accounts for the impact of external perturbations. By considering factors like voltage rise time and current fall time, alongside their reliability and weights, the model provides a more robust and adaptable assessment of battery health, even under fluctuating operating conditions.

What Problem Does This Paper Address?

Lithium-ion batteries are ubiquitous, powering everything from our phones to electric vehicles. However, their health degrades over time, impacting performance and safety. Accurately assessing this health in real-time is crucial. Existing methods often struggle with the complexity of battery systems, the influence of external factors (perturbations), and the need for interpretable results. This paper aims to bridge this gap by developing a health assessment model that is not only accurate but also resilient to perturbations, providing a clearer picture of a battery's true reliability.

Methods

The core of this work is the Evidential Reasoning (ER) rule, a framework for fusing uncertain information from multiple sources. The authors adapt this rule for battery health assessment by:

  • Selecting Key Indicators: They focus on the voltage rise time (Tvr) and current fall time (Tcf) during charging and discharging cycles, arguing these are accessible and informative indicators of battery health.
  • Calculating Evidence Parameters: For each indicator, they determine:
  • Reliability (r): Using a distance-based method to quantify how much the indicator's data fluctuates.
  • Weight (w): Employing the Coefficient of Variation (COV) method to reflect the indicator's importance.
  • Belief Distribution: Transforming raw indicator data into a belief structure across predefined health levels (high, medium, low) using a rule-based approach.
  • Health Assessment: The ER rule then aggregates these parameters to produce a comprehensive health status for the battery.
  • Perturbation Analysis: This is a key novelty. The authors introduce a method to quantify the impact of external perturbations (like temperature or current fluctuations) on battery health. They define perturbation intensity (σ) and perturbation variable (Δt) to simulate these disturbances. A perturbation coefficient (S) is then calculated to measure the battery's adaptability to these perturbations, indicating its reliability under stress.

Datasets

The study utilizes a publicly available dataset from the National Aeronautics and Space Administration (NASA) Prognostics Center of Excellence. This dataset contains aging data for an 18,650 lithium-ion battery, specifically the B0006 battery, collected over 167 charge-discharge cycles.

Key Insights and Ideas

  • Voltage and Current Timing as Health Proxies: The paper highlights that simple measurements like the time it takes for voltage to rise or current to fall during specific charging phases can be surprisingly effective indicators of a battery's overall health.
  • ER Rule's Robustness: The Evidential Reasoning rule proves adept at integrating diverse and uncertain information (indicator reliability, weights, belief distributions) into a coherent health assessment, offering a more nuanced view than single-metric approaches.
  • Quantifying Perturbation Impact: The introduction of a perturbation coefficient is a significant step towards understanding how real-world environmental and operational stresses affect battery reliability, moving beyond static health assessments.
  • Adaptability to Perturbations: The model doesn't just assess current health but also quantifies how well a battery can handle external disturbances, a crucial aspect for safety-critical applications.

Key Results

  • The proposed health assessment model, using voltage rise time and current fall time, effectively tracks the degradation of the lithium-ion battery over its lifecycle, aligning with observed capacity loss.
  • The perturbation analysis demonstrates that as perturbation intensity increases, the battery's reliability (as measured by the perturbation coefficient) decreases, indicating a reduced ability to withstand stress.
  • The model showed good performance even when an additional indicator, the maximum temperature gradient, was included, suggesting its flexibility.
  • A comparison with the Analytic Hierarchy Process (AHP) indicated that while AHP can assess overall health, the ER model is better at capturing the dynamic nature of battery deterioration over time.

Limitations and Future Work

  • Perturbation Measurement: The authors acknowledge that accurately measuring perturbation intensity in real-world engineering scenarios is challenging. Developing more precise methods for this is identified as a key area for future research.
  • Generalizability: While the paper suggests the model can be migrated to other battery types (aqueous, hybrid, solid-state), further validation on these specific chemistries is needed.
  • Subjectivity in AHP: The paper notes that AHP can be influenced by subjective factors, leading to potential inaccuracies, which their ER model aims to mitigate.

The proposed method offers a promising avenue for more reliable and adaptable battery health monitoring, especially in dynamic and demanding environments.

Summary updated

Abstract

Lithium-ion batteries are widely used in energy storage, small electronic devices and other fields due to their advantages of high energy density and long life cycles, as well as causing less damage to the environment than alternatives. For safety, it is essential to propose reasonable methods to assess batteries’ health statuses. Therefore, a health assessment model based on the evidential reasoning (ER) rule is proposed in this article. Firstly, the voltage rise time and the current fall time are taken as observation indicators, which contain information about the health status of lithium-ion batteries. Secondly, the information of various indicators is integrated into a belief structure, and the indicator reliability and indicator weights are adequately considered in the assessment model. Thirdly, there are some perturbations that will affect the operating status of batteries and cause the batteries’ reliability to fluctuate, so we use perturbation analysis to determine the adaptability of batteries to perturbations. We set two bounded parameters, the perturbation coefficient and the maximum perturbation error, to assess the reliability of lithium-ion batteries when experiencing perturbations. Finally, on the basis of the whole-life open data set of lithium-ion batteries from the National Aeronautics and Space Administration’s Prognostics Center of Excellence, the validity of the health assessment model and perturbation analysis is demonstrated.
BibTeX
@inproceedings{zhang2023new,
  author = {Yunyi Zhang and Shuchang Wang and Wei He and Wei Zhang and Shuaiwen Tang and Guohui Zhou},
  title = {{A New Health Analysis Method for Lithium-Ion Batteries Based on the Evidential Reasoning Rule Considering Perturbation}},
  booktitle = {Batteries},
  year = {2023},
  doi = {10.3390/batteries9020088},
}
Figure