Machine learning approach for full impedance spectrum study of Li-ion battery

Chen, Cuili; Yesilbas, Göktug; Lenz, Alexander; Schneider, Oliver; Knoll, Alois

doi:10.1109/iecon43393.2020.9254622

Cited by 4 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where the data point is f i and f ̇i is the slope at the value of x i for 1 ≤ i ≤ n. (3) Makima interpolation: Makima is a technology evolved from Akima. 35 It is a method based on local interpolation. Makima interpolation adopts the same method as Akima but adds additional elements to ensure that the value of the slope end area is close to zero.…”

Section: Discrete Interpolation Fitting Methodmentioning

confidence: 99%

Characteristic Investigation on Fluid Signals Based on a Combination of Empirical Mode Decomposition and Hilbert Transform in a Jet Impact-Negative Pressure Deamination Reactor

Zhang,

Chai,

Qiu

et al. 2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

The jet impact-negative pressure reactor (JI-NPR) is a novel wastewater treatment technology developed for the efficient removal of high-concentration ammonia nitrogen. However, the complex and transient nature of the flow behavior within the JI-NPR poses significant challenges for understanding the underlying fluid dynamics. In this work, a comprehensive signal-processing framework was developed to elucidate the flow characteristics inside the JI-NPR. First, a flow signal acquisition platform was established to capture the negative pressure signals during the treatment process. The empirical mode decomposition (EMD) technique was then employed to decompose the turbulent flow signals into a series of intrinsic mode functions (IMFs), representing multiscale turbulent eddy characteristics. To mitigate the effects of local noise and abrupt changes, various curve fitting methods, including cubic spline interpolation, piecewise cubic Hermite interpolating polynomial, and Makima interpolation, were utilized to smooth the IMF signals. The Hilbert transform was subsequently applied to extract the instantaneous frequency features of the smoothed IMFs, enabling more accurate quantification of the nonstationary and nonlinear flow behavior. The results revealed that the low-frequency IMFs were associated with the interactions between the wastewater jet and negative pressure, while the highfrequency IMFs reflected the internal dynamic evolution of the fluid. Furthermore, the multiple regression analysis approach was adopted to quantify the relationship between the IMF feature parameters and the critical performance metric of the denitrification efficiency. The decision tree regression model was identified as a particularly suitable technique, as it can flexibly capture both linear and nonlinear dependencies and effectively identify the most influential variables. This integrated approach of EMD, curve fitting, Hilbert transform, and regression analysis methods provides valuable insights into the quantitative impact of multiscale turbulent eddies on the overall performance of the JI-NPR system. These findings are expected to guide targeted optimization of the reactor design to enhance the denitrification efficiency, a crucial goal for the practical application of this wastewater treatment technology.

show abstract

Section: Discrete Interpolation Fitting Methodmentioning

confidence: 99%

Characteristic Investigation on Fluid Signals Based on a Combination of Empirical Mode Decomposition and Hilbert Transform in a Jet Impact-Negative Pressure Deamination Reactor

Zhang,

Chai,

Qiu

et al. 2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…Psuedo-random sequences were also employed in [6] with a focus on various implementation aspects in EV. Authors in [7] proposed using machine learning to reconstruct the entire impedance spectrum based on a few multi-sine signal injection measurements, which also reduced the execution time significantly. In [8], a very comprehensive analysis of battery ageing in different storage conditions was performed on long-term data.…”

Section: State-of-the-art Reviewmentioning

confidence: 99%

Capacity State-of-Health Estimation of Electric Vehicle Batteries Using Machine Learning and Impedance Measurements

et al. 2022

View full text Add to dashboard Cite

With the increasing adoption of electric vehicles (EVs) by the general public, a lot of research is being conducted in Li-ion battery-related topics, where state-of-health (SoH) estimation has a prominent role. Accurate knowledge of this parameter is essential for efficient and safe EV operation. In this work, machine-learning techniques are applied to estimate this parameter in EV applications and in diverse scenarios. After thoroughly analysing cell ageing in different storage conditions, a novel approach based on impedance data is developed for SoH estimation. A fully-connected feed-forward neural network (FC-FNN) is employed to estimate the battery’s maximum available capacity from a small set of impedance measurements. The method was tested for estimation in long-term scenarios and for diverse degradation procedures with data from real EV batteries. High accuracy was obtained in all situations, with a mean absolute error as low as 0.9%. Thus, the proposed algorithm constitutes a powerful and viable solution for fast and accurate SoH estimation in real-world battery management systems.

show abstract

“…Makima is a technique that evolved from Akima [30]. Originally, Akima was based on a cubic polynomial.…”

Section: Makima Interpolationmentioning

confidence: 99%

Spatial interpolation method comparison for physico-chemical parameters of river water in Klang River using MATLAB

et al. 2022

View full text Add to dashboard Cite

Water quality is one of the most highly debated issues worldwide at the moment. Inadequate water supplies affect human health, hinder food production, and degrade the environment. Using contemporary technology to analyze pollution statistics can help solve pollution issues. One option is to take advantage of advancements in intelligent data processing to conduct hydrological parameter analysis. To perform conclusive water quality studies, a lot of data is necessary. Unfilled data (information gaps) in the long-term hydrological data set may be due to equipment faults, collection schedule delays, or the data collection officer’s absence. The lack of hydrological data skews its interpretation. Therefore, interpolation is used to recreate and fill missing hydrological data. From 2012 to 2017, the Klang River’s biochemical oxygen demand (BOD) in Selangor, Malaysia, was sampled. This study examined three methods of interpolation for their effectiveness using the MATLAB software: piecewise cubic hermite interpolating polynomial (PCHIP), cubic Spline data interpolation (Spline), and modified Akima partitioned cubic hermite interpolation (Makima). The accuracy is assessed using root mean square error (RMSE). All interpolation algorithms offer excellent results with low RMSE. However, PCHIP delivers the best match between interpolated and original data.

show abstract

Machine learning approach for full impedance spectrum study of Li-ion battery

Cited by 4 publications

References 14 publications

Characteristic Investigation on Fluid Signals Based on a Combination of Empirical Mode Decomposition and Hilbert Transform in a Jet Impact-Negative Pressure Deamination Reactor

Characteristic Investigation on Fluid Signals Based on a Combination of Empirical Mode Decomposition and Hilbert Transform in a Jet Impact-Negative Pressure Deamination Reactor

Capacity State-of-Health Estimation of Electric Vehicle Batteries Using Machine Learning and Impedance Measurements

Spatial interpolation method comparison for physico-chemical parameters of river water in Klang River using MATLAB

Contact Info

Product

Resources

About