State-of-charge estimation of lithium-ion batteries using LSTM and UKF

Yang, Fangfang; Zhang, Shaohui; Li, Weihua; Miao, Qiang

doi:10.1016/j.energy.2020.117664

Cited by 274 publications

(84 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [18], a state-of-charge battery ESS prediction model was proposed by combining LSTM and an unscented Kalman filter. The authors in [19] proposed a data-driven model for handling the EV demand uncertainty. Their results established that the data-driven model can provide a superior framework for handling uncertainty owing to its highly effective memory unit in the network.…”

Section: B Literatrue Reviewmentioning

confidence: 99%

“…The fitness function of the ELPSO algorithm can be expressed as (19) To protect the forecasting model from overfitting, the fitness function includes the training and validation sample error. Here, the weights of each sample are assigned as 0.5.…”

Section: Enhanced Learning Particle Swarm Optimizationmentioning

confidence: 99%

“…Here, the weights of each sample are assigned as 0.5. According to (11)- (19), the optimal GRU model can be determined via repeated iterations.…”

Section: Enhanced Learning Particle Swarm Optimizationmentioning

confidence: 99%

See 2 more Smart Citations

Data-Driven Bidding Strategy for DER Aggregator Based on Gated Recurrent Unit–Enhanced Learning Particle Swarm Optimization

2021

View full text Add to dashboard Cite

Distributed energy resources (DERs) such as wind turbines (WTs), photovoltaics (PVs), energy storage systems (ESSs), local loads, and demand response (DR) are highly valued for environmental protection. However, their volatility poses several risks to the DER aggregator while formulating a profitable strategy for bidding in the day-ahead power market. This study proposes a data-driven bidding strategy framework for a DER aggregator confronted with various uncertainties. First, a data-driven forecasting model involving gated recurrent unit-enhanced learning particle swarm optimization (GRU-ELPSO) with improved mutual information (IMI) is employed to model renewables and local loads. It is critical for a DER aggregator to accurately estimate these components before bidding in the day-ahead power market. This aids in reducing the penalty costs of forecasting errors. Second, an optimal bidding strategy that is based on the information gap decision theory (IGDT) is formulated to address market price uncertainty. The DER aggregator is assumed to be risk-averse (RA) or risk-seeker (RS), and the corresponding bidding strategies are formulated according to the risk preferences thereof. Thenr, an hourly bidding profile is created for the DER aggregator to bid successfully in the day-ahead power market. The proposed datadriven bidding framework is evaluated using an illustrative system wherein a dataset is obtained from the PJM market. The results reveal the effectiveness of handling uncertainty by providing accurate forecasting results. In addition, the DER aggregator can bid effectively in the day-ahead power market according to its preference for robustness or high profit, with a suitable bidding profile.

show abstract

Section: B Literatrue Reviewmentioning

confidence: 99%

Section: Enhanced Learning Particle Swarm Optimizationmentioning

confidence: 99%

See 1 more Smart Citation

Data-Driven Bidding Strategy for DER Aggregator Based on Gated Recurrent Unit–Enhanced Learning Particle Swarm Optimization

2021

View full text Add to dashboard Cite

show abstract

“…The LSTM cell nucleus is used to replace the cell nucleus of the traditional dynamic neural network; therefore, it has long-term memory capabilities. The information forgotten by the LSTM cell nucleus feeds important information to the hidden layer, including the forgetting gate, input gate, and output gate [38]. The memory cell of an LSTM recurrent neural network is shown in Fig.…”

Section: B Lstm Neural Networkmentioning

confidence: 99%

State of Charge Estimation of Lithium-Ion Batteries Using LSTM and NARX Neural Networks

Wei

Min

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Highly accurate state of charge (SOC) estimation of lithium-ion batteries is one of the key technologies of battery management systems in electric vehicles. The performance of SOC estimation directly influences the driving range and safety of these vehicles. Due to external disturbances, temperature variation and electromagnetic interference, accurate SOC estimation becomes difficult. To accurately estimate the SOC of lithium-ion batteries, this paper presents a novel machine-learning method to address the risk of gradient explosion and gradient decent using the dynamic nonlinear auto-regressive models with exogenous input neural network (NARX) with long short-term memories (LSTM).The proposed hybrid NARX model embeds LSTM memory, which provides jump-ahead connections in the time-unfolded model. These jump-ahead connections provide a shorter path for the propagation of gradient information, therefore reducing long-term dependence on the recurrent neural network. Experimental results show that the estimation performance root mean square error (RMSE) of the proposed model is less than 1%, and this model has better multitime prediction performance. Finally, the hybrid NARX and LSTM model is compared with the standard back propagation neural network based on particle swarm optimization (BPNN-PSO), the least-squares support vector machine (LS-SVM) and LSTM existing models under urban dynamometer driving schedule (UDDS) and dynamic stress test (DST) conditions. The proposed hybrid NARX-LSTM model yield relative to other methods and can estimate the battery SOC with high accuracy. The RMSE of proposed model is improved by approximately 60% compared with the standard LSTM INDEX TERMS electric vehicles; state of charge; lithium-ion batteries; NARX; LSTM

show abstract

“…The open circuit voltage method can accurately reflect the battery SOC, but it takes a long time to measure the open circuit voltage (OCV) accurately [7][8][9][10][11]. The equivalent circuit model method mainly includes a series of Kalman filter derivative methods, such as: extended Kalman filter (EKF), unscented Kalman filter (UKF), and particle filtering (PF) [12][13][14]. The accuracy of the Kalman filter method depends on an accurate battery model which is difficult to obtain [15][16][17].…”

Section: Introductionmentioning

confidence: 99%

State of Charge Estimation for Lithium-Ion Battery Based on NARX Recurrent Neural Network and Moving Window Method

Wang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

An accurate state of charge (SOC) estimation depends on an accurate battery model. The influence of nonlinear and unstable interference factors makes the accurate SOC estimation difficult. To obtain an accurate battery model, a method based on the NARX (nonlinear autoregressive network with exogenous inputs) recurrent neural network and moving window method is proposed. This paper improves the accuracy, modelling speed and robustness of SOC estimation from the following three aspects. First, to overcome the excessive reliance on the amount of data in the model training process, the NARX recurrent neural network is used to establish the battery model. NARX (nonlinear autoregressive with external input) recurrent neural network with the delay and feedback functions can keep the input and output of a previous moment and add it to the calculation of the next moment. Therefore, better estimation results are achieved using a small amount of data; second, the moving window method is used against the gradient explosion and the gradient vanishing that may occur in the NARX model training process. Third, by comparing it with other methods under different working conditions and different temperatures, the validity of the proposed model is verified. The results indicate that the proposed model has a higher accuracy and speed of the SOC estimation. The RMSE performance of the proposed model is reduced by approximately 65%, and the execution time is shortened by approximately 50%.

show abstract

State-of-charge estimation of lithium-ion batteries using LSTM and UKF

Cited by 274 publications

References 25 publications

Data-Driven Bidding Strategy for DER Aggregator Based on Gated Recurrent Unit–Enhanced Learning Particle Swarm Optimization

Data-Driven Bidding Strategy for DER Aggregator Based on Gated Recurrent Unit–Enhanced Learning Particle Swarm Optimization

State of Charge Estimation of Lithium-Ion Batteries Using LSTM and NARX Neural Networks

State of Charge Estimation for Lithium-Ion Battery Based on NARX Recurrent Neural Network and Moving Window Method

Contact Info

Product

Resources

About