On-line WSN SoC estimation using Gaussian Process Regression: An Adaptive Machine Learning Approach

Ali, Omer; Ishak, Mohamad Khairi; Ahmed, Ashraf Bani; Salleh, Mohd Khairul Mohd; Ooi, Chia Ai; Akbar, Muhammad Firdaus; Khan, Imran

doi:10.1016/j.aej.2022.02.067

Cited by 29 publications

(9 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gaussian regression is also more suitable for lightweight data processing. Ali et al improved the adaptive Gaussian-regression-based SOC estimate technique, which directly maps battery parameters, for instance, temperature, capacity, and voltage, to the corresponding model, achieving SOC estimation on embedded platforms [59]. A Gaussian process regression (GPR)-based data-driven approach is proposed to address the issue of inconsistency in battery packs, which results in an SOC estimation performance lower than that of the conventional models with estimation errors of 3.9% under various dynamic cycles, temperatures, aging circumstances, and even extreme situations [60].…”

Section: Data-driven Approachmentioning

confidence: 99%

Advances in the Study of Techniques to Determine the Lithium-Ion Battery’s State of Charge

Liu,

Gao,

Marma

et al. 2024

Energies

View full text Add to dashboard Cite

This study explores the challenges and advances in the estimation of the state of charge (SOC) of lithium-ion batteries (LIBs), which are crucial to optimizing their performance and lifespan. This review focuses on four main techniques of SOC estimation: experimental measurement, modeling approach, data-driven approach, and joint estimation approach, highlighting the limitations and potential inaccuracies of each method. This study suggests a combined approach, incorporating correction parameters and closed-loop feedback, to improve measurement accuracy. It introduces a multi-physics model that considers temperature, charging rate, and aging effects and proposes the integration of models and algorithms for optimal estimation of SOC. This research emphasizes the importance of considering temperature and aging factors in data-driven approaches. It suggests that the fusion of different methods could lead to more accurate SOC predictions, an important area for future research.

show abstract

Section: Data-driven Approachmentioning

confidence: 99%

Advances in the Study of Techniques to Determine the Lithium-Ion Battery’s State of Charge

Liu,

Gao,

Marma

et al. 2024

Energies

View full text Add to dashboard Cite

show abstract

“…The SOC estimation methods proposed in the previous literatures are classified into three categories: the direct measurement [5][6][7] methods, the estimation methods [8][9][10][11] with a data-driven predictive model, and the estimation methods 7,[12][13][14][15][16] based on the battery models. Based on the directly measurable battery variables, the direct measurement methods mainly include the coulomb counting method, 7 open circuit voltage (OCV) method, 5 impedance spectroscopy method, 6 and internal resistance method.…”

Section: Introductionmentioning

confidence: 99%

“…With the data‐driven predictive model, the SOC estimation methods mainly include neural networks, 9 support vector machines, 10 and deep learning 11 . Based on a convolution network, Tian et al 11 proposed a method that requires the short‐term data to estimate SOC and state of health (SOH).…”

Section: Introductionmentioning

confidence: 99%

State of charge estimation for the vanadium redox flow battery based on the Sage–Husa adaptive extended Kalman filter

Zixuan

Fan

et al. 2023

Circuit Theory & Apps

View full text Add to dashboard Cite

SummaryFocusing on improving the accuracy of vanadium redox flow battery (VRFB) state of charge (SOC) estimation, this paper will combine the extended Kalman filter (EKF) estimator with the Sage–Husa adaptive method, referred to as the SAEKF estimator. Firstly, a second‐order Equivalent Circuit Model (ECM) is established to provide the correct parameterization and degrees of freedom. Next, under a hybrid pulse power characterization (HPPC) text, the parameter estimator in MATLAB/Simulink is utilized to identify the ECM parameters by using nonlinear least squares method with Trust‐Region‐Reflective (TRR) iterative algorithm. Then, during the HPPC test, the estimators implemented on the ARM‐Cortex STM32F103 estimate the VRFB SOC under both the 0°C and 5°C conditions. Finally, the mean absolute error (MAE) of the IEKF estimator are 4.66% at the 0°C and 0.895% at the 5°C. In the meantime, the MAE of the SAEKF estimator are 4.15% at the 0°C and 0.703% at the 5°C. The evaluation factors of the 0°C and 5°C experiments in the SAEKF estimator are smaller than those of the IEKF and EKF estimators, indicating that the SAEKF estimator outperforms the IEKF estimator in VRFB SOC estimation.

show abstract

“…Figure 10: Evaluation of the diagnostic results of BRB-AAW in comparison to BRBUpon conducting a comparison between the outcomes of BRB-AAW and BRB, it becomes apparent that BRB-AAW's values more accurately reflect the real scenario, particularly when the standard value is 0. The test results indicated that BRB-AAW's overall accuracy, narrow neural networks, and SVM were 92.25%, 90.29%, and 87.98%, respectively, while the corresponding values for BRB were 90.24%, 23.35%, and 1.09%[30]. The scattered diagnostic results of the BRB approach can be attributed to the use of static attribute weights, which can lead to incorrect fault type identification based on attribute weights.…”

mentioning

confidence: 99%

A Fault Diagnosis Method for Wireless Sensor Network nodes based on Belief Rule Base with Adaptive Attribute Weights

Shi

Sun

et al. 2023

Preprint

View full text Add to dashboard Cite

Due to wireless sensor networks (WSNs)’ harsh operating environment and ultralong operating hours, node failures are inevitable. Ensuring the dependable collection of data necessitates the utmost importance of diagnosing faults in nodes within the wireless sensor network. Typically, the initial step in the fault diagnosis of WSN nodes involves extracting numerical features from neighboring nodes. However, a solitary data feature assigned a high weight fails to effectively discriminate between all types of faults. Therefore, this study introduces an enhanced variant of the traditional belief rule base (BRB), called the belief rule base with adaptive attribute weights (BRB-AAW). First, the data features are extracted as input attributes for the model. Second, a fault diagnosis model for WSN nodes, incorporating BRB-AAW, is established by integrating expert knowledge initialization parameters with the extracted data features. To optimize the model's initial parameters, the projection covariance matrix adaptive evolution strategy (P-CMA-ES) algorithm is employed. Finally, a comprehensive case study is designed to verify the accuracy and effectiveness of the proposed method. The outcomes of the case study indicate a discernible enhancement in the accuracy of WSN node fault diagnosis when compared to the traditional BRB method.

show abstract

On-line WSN SoC estimation using Gaussian Process Regression: An Adaptive Machine Learning Approach

Cited by 29 publications

References 61 publications

Advances in the Study of Techniques to Determine the Lithium-Ion Battery’s State of Charge

Advances in the Study of Techniques to Determine the Lithium-Ion Battery’s State of Charge

State of charge estimation for the vanadium redox flow battery based on the Sage–Husa adaptive extended Kalman filter

A Fault Diagnosis Method for Wireless Sensor Network nodes based on Belief Rule Base with Adaptive Attribute Weights

Contact Info

Product

Resources

About