A rapid screening and regrouping approach based on neural networks for large-scale retired lithium-ion cells in second-use applications

Lai, Xin; Qiao, Dongdong; Zheng, Yuejiu; Ouyang, Minggao; Han, Xiao; Zhou, Long

doi:10.1016/j.jclepro.2018.12.210

Cited by 105 publications

(33 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, other sub-models are including the SOC calculation, voltage, internal resistance, Coulomb efficiency, capacity fade and self-discharge. 21 The 108 single-cell models are connected in series to form a battery system which is shown in Figure 2.…”

Section: Battery Modelmentioning

confidence: 99%

“…The Back Propagation Neural Network (BPNN)model was built by the charging voltage variation and capacity to achieve the fast classification of different capacity cells. 21 We further proposed discharge in series to rapidly classify the cells after each cell is fully charged, and used Genetic Algorithm and back-propagation (GA-BP) neural network for model training and capacity estimation. 22 In those works, according to capacity and internal resistance, the cells were simply divided during the regrouping.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A novel fast estimation and regroup method of retired lithium‐ion battery cells

et al. 2020

Self Cite

View full text Add to dashboard Cite

With the commercialization of the electric vehicles, the large-scale lithium-ion cells as the power of electric vehicles are to be retired. The second-use of retired cells is of great significance to improve the battery economy. A fast classification and regroup evaluation method of the retired lithium-ion cells are proposed in this paper to improve the classification efficiency of retired lithium-ion cells and adapt to the regroup under different conditions. The lithium-ion cells after being balanced in parallel are charged in series with a constant current. A support vector regression (SVR) model with the parameters optimized by the particle swarm optimization (PSO) algorithm is built for the fast capacity estimation and the error will not exceed 0.3%. Different cells regrouped means different performance. In order to improve the consistency of retired cells and satisfy different using conditions, a Weighted-K-means algorithm is proposed in this paper to regroup the cells with the known capacity and internal resistance. The classification method is evaluated by the voltage consistency of cells using different working conditions, which indicates capacity occupied a large proportion can meet the requirement of energy condition meanwhile keep a good consistency. But the resistance will dominate in algorithm under conditions which have requirement for instantaneous power.

show abstract

Section: Battery Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A novel fast estimation and regroup method of retired lithium‐ion battery cells

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Nowadays, battery manufacturers and governments are facing great pressure to recycle and dispose of batteries, because a large number of lithium-ion batteries are retired from EVs for safe operation and longer driving range after being employed in EVs for a few years [5]. However, the retired lithium-ion batteries' capacity is 70% to 80% of their initial capacity, and the certain residual capacity can be used for energy storage in an electricity grid after they are tested, selected, and classified [6,7]. The accumulatively installed capacity of electrochemical energy storage projects reached 1072.7 MW in China by the end of 2018, and out of all electrochemical energy storage projects in China, the quotient of lithium-ion batteries was maximal and achieved 70.7% [8].…”

Section: Introductionmentioning

confidence: 99%

A Layered Bidirectional Active Equalization Method for Retired Power Lithium-Ion Batteries for Energy Storage Applications

et al. 2020

View full text Add to dashboard Cite

The power from lithium-ion batteries can be retired from electric vehicles (EVs) and can be used for energy storage applications when the residual capacity is up to 70% of their initial capacity. The retired batteries have characteristics of serious inconsistency. In order to solve this problem, a layered bidirectional active equalization topology is proposed in this paper. Specifically, a bridge-type equalization topology based on an inductor is adopted in the bottom layer, and the distributed equalization topological structure based on the bidirectional BUCK-BOOST circuit is adopted in the top layer. State of charge (SOC) is used as the equalization target variable, and the bottom layer equalization algorithm based on a “partition” idea and route optimization is proposed. The static equalization experiments and charge equalization experiments are performed by the 12 retired batteries selected from an electric sanitation vehicle. The results show that the proposed equalization method can reduce the SOC difference between retired batteries and can effectively improve the inconsistency of the retired battery pack with a faster equalization speed.

show abstract

“…47 The temperature-compensated models were conducted by using the Extended Kalman Filter (EKF), which were used as the implantable charger. 52 Therefore, it is necessary to construct the adaptive model based on the internal mechanism theory, which can achieve the optimum life for f lithium-ion battery packs, protecting the instantaneous power supply capacity and improving its energy utilization. 49 The electrochemical model was conducted by using the charging optimization 50 and the dynamic model was investigated incorporating electro-thermal and aging aspects.…”

Section: Introductionmentioning

confidence: 99%

“…51 A rapid screening and regrouping approach was proposed to manage the large-scale retired lithium-ion battery cells in second-use applications by using the neural network algorithm. 52 Therefore, it is necessary to construct the adaptive model based on the internal mechanism theory, which can achieve the optimum life for f lithium-ion battery packs, protecting the instantaneous power supply capacity and improving its energy utilization.…”

Section: Introductionmentioning

confidence: 99%

An improved packing equivalent circuit modeling method with the cell‐to‐cell consistency state evaluation of the internal connected lithium‐ion batteries

Wang

Shi

Fernandez

et al. 2019

Energy Science & Engineering

View full text Add to dashboard Cite

The existing equivalent modeling methods reported in literature focuses mainly on the battery cells and do not take the packing consistency state into consideration, which exists on the internal connected cells of the lithium‐ion battery pack. An improved equivalent circuit model is constructed and reported in this manuscript for the first time, which can be used to characterize the working characteristics of the packing lithium‐ion batteries. A new equilibrium concept named as state of balance is proposed as well as the calculation process, which is realized by considering the real‐time detected internal battery cell voltages. In addition, this new equilibrium concept aims to obtain more information on the real‐time consistency characterization of the battery pack. The improved adaptive equivalent circuit model is investigated by using the improved splice modeling method, in which the statistical noise properties are corrected and the additional parallel resistance‐capacitance circuit is introduced. The parameter correction treatment is carried out by comparing the estimated and experimental detected closed circuit voltages. Furthermore, the tracking error is found to be 0.005 V and accounts for 0.119% of the nominal battery voltage. By taking the packing consistency state and temperature correction into consideration, the accurate working characteristic expression is realized in the improved equivalent circuit modeling process. Finally, the model proposed in this manuscript presents a great number of advantages compared to other methods reported so far, like has the high accuracy, and the ability to protect the security of the lithium‐ion battery pack in the power supply application.

show abstract

A rapid screening and regrouping approach based on neural networks for large-scale retired lithium-ion cells in second-use applications

Cited by 105 publications

References 41 publications

A novel fast estimation and regroup method of retired lithium‐ion battery cells

A novel fast estimation and regroup method of retired lithium‐ion battery cells

A Layered Bidirectional Active Equalization Method for Retired Power Lithium-Ion Batteries for Energy Storage Applications

An improved packing equivalent circuit modeling method with the cell‐to‐cell consistency state evaluation of the internal connected lithium‐ion batteries

Contact Info

Product

Resources

About