An Improved Gated Recurrent Unit Neural Network for State-of-Charge Estimation of Lithium-Ion Battery

Chen, Jianlong; Lu, Chenlei; Chen, Cong; Cheng, Hangyu; Xuan, Dongji

doi:10.3390/app12052305

Cited by 21 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, due to the heavy mathematical processing load of the LSTM method, the GRU method has emerged as a simpler method. The GRU method is one of the most used methods in battery SOC estimation as it consists of fewer gates and performs faster processing [18]. Additionally, the BiLSTM method, which is used in studies to estimate battery SOC, is based on the LSTM method and transfers information in both directions [19].…”

Section: Related Workmentioning

confidence: 99%

Performance Comparison of Lithium Polymer Battery SOC Estimation Using GWO-BiLSTM and Cutting-Edge Deep Learning Methods

Taş

Bal

Uysal

2023

Preprint

View full text Add to dashboard Cite

In this study, the GWO-BiLSTM method has been proposed by successfully estimating the SOC with the BiLSTM deep learning method using the hyper-parameter values determined by the GWO method of the lithium polymer battery. In studies using deep learning methods, it is important to solve the problems of underfitting, overfitting, and estimation error by determining the hyper-parameters appropriately. EV, HEV, and robots are used more healthily with the successful, reliable, and fast SOC estimation, which has an important place in the Battery Management System. The success of the proposed method was verified by comparing the cutting-edge data-based deep learning methods and the BiLSTM method with the SOC estimation MAE, MSE, RMSE, and Runtime(s) metrics. In the comparison, the prediction successes of the BiLSTM method, which was trained with the optimal hyper-parameter values obtained by the GWO method, with the cutting-edge deep learning methods trained with the hyper-parameter values obtained through trial and error were compared. The GWO-BiLSTM method was the most successful method with RMSE of 0.09244% and R2 of 0.9987 values according to the average results of SOC estimation made with the lithium polymer battery data set, which was created by experiments performed at different discharge levels and is new in the literature.

show abstract

Section: Related Workmentioning

confidence: 99%

Performance Comparison of Lithium Polymer Battery SOC Estimation Using GWO-BiLSTM and Cutting-Edge Deep Learning Methods

Taş

Bal

Uysal

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Unlike other neural networks, the RNN network consists of some copies of a unique unit that are connected like a chain, this structure allows Simple RNN network suffers from gradient vanishing and gradient explosion, in order to solve these problems LSTM and GRU are established [11,12]. In compression with LSTM, the GRU cell has a simpler structure due to one less gate [28]. Therefore, the number of parameters under the same network structure in GRU is less than in LSTM, which can reduce the risk of overfitting and boost the convergence rate while its performance is just like LSTM cell [18].…”

Section: Gru Algorithmmentioning

confidence: 99%

“…4: a recurrent neural network schematic GRU architecture is shown in Fig. 4, in compression with simple RNN, the GRU-RNN has a relevance gate which is responsible for controlling the retention of historical information, and an update gate which controls the effect of the previous cell state and the current input on the new cell state [28]. information from the current input and Information from the previous hidden state which can be considered as the memory of the network is passed through the sigmoid function, so the Values come out between 0 and 1.…”

Section: Gru Algorithmmentioning

confidence: 99%

online state of charge estimation of lithium-ion batteries using gated recurrent unit neural network and a least mean square adaptive filter

Radmanesh

rad

2022

Preprint

View full text Add to dashboard Cite

State of charge (SOC) estimation of lithium-ion batteries is the most important role of a battery management system. to improve the SOC estimation speed and accuracy in the operational environment, a novel method is proposed by combining a gated recurrent unit (GRU) neural network and a least mean square (LMS) adaptive filter. First a GRU network is used to estimate the SOC based on the battery measurement data. Then the LMS filter is used for online error reduction through unpredicted operation conditions, the LMS is a lite adaptive filter that updates its coefficient based on operation conditions with low computation cost. To verify the method robustness, its performance was checked under constant and varying temperature for standard drive cycles like UDDS and LA92. The proposed method is able to estimate SOC with less than 10-min discharge voltage, current and temperature data as the input with an error of less than 0.6\% during working hour. Therefore, compared with conventional methods like LSTM and GRU the proposed GRU-LMS method has better speed and accuracy in the SOC estimation.

show abstract

“…The authors used GA to optimize the number of layers and neurons in the GRU model structure. The results showed that this method is highly accurate and robust [32]. Zhang et al combined GA with a fuzzy logic control neural network algorithm, utilizing the neural network to accurately estimate the dynamic SOC of the battery based on the static SOC.…”

Section: Introductionmentioning

confidence: 99%

Influence of Time-Series Length and Hyperparameters on Temporal Convolutional Neural Network Training in Low-Power Battery SOC Estimation

Wang,

Lu,

2023

Applied Sciences

View full text Add to dashboard Cite

In battery management systems, state of charge (SOC) estimation is important for ensuring the safety and reliability of batteries. Currently, there are various methods for estimating SOC, and the neural network method is the most popular. However, when the battery’s SOC is low (below 20%), the uncertainty in neural network parameters can lead to significant bias in SOC estimation. To address these problems, this study proposes a method based on genetic algorithm (GA) optimization of a time-serialization convolutional neural network (TSCNN) model. First, the population is initialized according to the optimized hyperparameters of the TSCNN model, whereby the experimental data are converted into time-series data. Subsequently, neural network models are built based on the population, thereby using the effect of the network as the fitness function for GA optimization. Finally, an optimized network structure is obtained for accurate SOC estimation. During the optimization process, the optimized data exhibited abnormal phenomena, usually manifested as exceeding the data limits or being zero. In the past, abnormal data were discarded and new data were regenerated; however, this reduces the correlation between data. Therefore, this study proposes a check function to enhance the correlation between the data, converting abnormal data into normal data by limiting the data range. To the best of our knowledge, it is the first time that a GA is being proposed to optimize the time-series length of a convolutional neural network (CNN) while the neural network parameters are optimized so that the time-series length and neural network parameters achieve the best match. In the experimental results, the maximum error was 4.55% for the dynamic stress test (DST) dataset and 2.58% for the urban dynamometer driving schedule (UDDS) dataset. When the battery SOC was below 20%, the estimation error did not incur a huge error. Therefore, the optimization method proposed for the TSCNN model in this study can effectively improve the accuracy and reliability of SOC estimation in the low-battery state.

show abstract

An Improved Gated Recurrent Unit Neural Network for State-of-Charge Estimation of Lithium-Ion Battery

Cited by 21 publications

References 28 publications

Performance Comparison of Lithium Polymer Battery SOC Estimation Using GWO-BiLSTM and Cutting-Edge Deep Learning Methods

Performance Comparison of Lithium Polymer Battery SOC Estimation Using GWO-BiLSTM and Cutting-Edge Deep Learning Methods

online state of charge estimation of lithium-ion batteries using gated recurrent unit neural network and a least mean square adaptive filter

Influence of Time-Series Length and Hyperparameters on Temporal Convolutional Neural Network Training in Low-Power Battery SOC Estimation

Contact Info

Product

Resources

About