A New Hybrid Neural Network Method for State-of-Health Estimation of Lithium-Ion Battery

Bao, Zhengyi; Jiang, Jiahao; Zhu, Chunxiang; Gao, Mingyu

doi:10.3390/en15124399

Cited by 22 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in some cases, feedback on the future sequence value at a certain time is considered when building the model. Thi information can be used to modify a model [41]. Therefore, a BiGRU model was con structed, as shown in Figure 1.…”

Section: Bidirectional Gate Recurrent Unitmentioning

confidence: 99%

Lithium-Ion Battery Remaining Useful Life Prediction Based on Hybrid Model

Tang

Wan

Wang³

et al. 2023

Sustainability

View full text Add to dashboard Cite

Accurate prediction of the remaining useful life (RUL) is a key function for ensuring the safety and stability of lithium-ion batteries. To solve the capacity regeneration and model adaptability under different working conditions, a hybrid RUL prediction model based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and a bi-directional gated recurrent unit (BiGRU) is proposed. CEEMDAN is used to divide the capacity into intrinsic mode functions (IMFs) to reduce the impact of capacity regeneration. In addition, an improved grey wolf optimizer (IGOW) is proposed to maintain the reliability of the BiGRU network. The diversity of the initial population in the GWO algorithm was improved using chaotic tent mapping. An improved control factor and dynamic population weight are adopted to accelerate the convergence speed of the algorithm. Finally, capacity and RUL prediction experiments are conducted to verify the battery prediction performance under different training data and working conditions. The results indicate that the proposed method can achieve an MAE of less than 4% with only 30% of the training set, which is verified using the CALCE and NASA battery data.

show abstract

Section: Bidirectional Gate Recurrent Unitmentioning

confidence: 99%

Lithium-Ion Battery Remaining Useful Life Prediction Based on Hybrid Model

Tang

Wan

Wang³

et al. 2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…These include OCV, model-based, data-driven, hybrid, 6 and combined methods. 7 In addition, various methods, such as the total leastsquares algorithm for capacity estimation, 8 hybrid neural networks, 9 and improved long short-term memory (LSTM) algorithms, 10 have been studied to estimate the SOH. 11,12 Accordingly, original equipment manufacturers manufacture eco-friendly vehicles that demand certain accuracies in terms of the SOC and SOH, that is, typically within 5% and 10% for the SOC and SOH, respectively.…”

Section: Introductionmentioning

confidence: 99%

Method for evaluating the accuracy of state‐of‐charge (SOC)/state‐of‐health (SOH) estimation of BMSs

Song,

Shin

2023

Energy Science & Engineering

View full text Add to dashboard Cite

The battery management system (BMS), which is widely adopted in various eco‐friendly automobiles that use batteries for propulsion, increases energy efficiency and extends the usable life by monitoring and optimally managing batteries. Eco‐friendly automobile original equipment manufacturers demand improvements in state‐of‐charge (SOC) and state‐of‐health (SOH) accuracies. However, because standards for evaluating accuracy do not exist, evaluations are performed using methods suggested by BMS developers or researchers, which differ from each other. Thus, the reliability of the SOC and SOH accuracy results is low. To accurately evaluate the SOC/SOH, it is necessary to reflect various conditions such as the power of the battery during actual vehicle operation and the temperature inside and outside the battery at the time. Therefore, a method is proposed herein that individually or simultaneously evaluates the SOC and SOH estimation accuracies of the BMS while considering an eco‐friendly vehicle's actual driving environment, instead of simple charge and discharge test conditions and room temperature. The reference values for evaluating the accuracy of SOC/SOH estimation in a battery can be defined as follows: Full charge and full discharge are conducted before applying the evaluation pattern to calculate the precise available capacity of the battery. Subsequently, it is fully charged and discharged to its initial SOC. A load pattern reflecting the load conditions of the subsequent application is repeatedly applied. After concluding the evaluation pattern, the remaining capacity is discharged, thereby determining the reference value. The proposed method is applied to two types of BMSs to evaluate its accuracy in SOC/SOH estimation and the validity of the proposed method was verified.

show abstract

Machine learning and neural network supported state of health simulation and forecasting model for lithium-ion battery

Qi,

Yan,

et al. 2023

Front. Energy

View full text Add to dashboard Cite

A New Hybrid Neural Network Method for State-of-Health Estimation of Lithium-Ion Battery

Cited by 22 publications

References 32 publications

Lithium-Ion Battery Remaining Useful Life Prediction Based on Hybrid Model

Lithium-Ion Battery Remaining Useful Life Prediction Based on Hybrid Model

Method for evaluating the accuracy of state‐of‐charge (SOC)/state‐of‐health (SOH) estimation of BMSs

Machine learning and neural network supported state of health simulation and forecasting model for lithium-ion battery

Contact Info

Product

Resources

About