State of Charge Estimation in Lithium-Sulfur Cells Using LSTM Recurrent Neural Networks

Abstract

This paper presents a framework for all-state estimation of Lithium-Sulfur (Li-S) battery cells based on a Long Short-Term Memory Recurrent Neural Network (LSTM RNN) model. Under the proposed framework, the LSTM RNN model is calibrated into the single task of State of Charge (SoC) estimation for fresh Li-S prototype cells. The Adaptive Moment Estimation (Adam) solver is used. Data sets for training and testing are derived from experiments using the WLTP duty cycles. The calibrated LSTM RNN structure is described for the purposes of training and testing with experimental datasets, so as to generate a network that can be deployed in real-time system. The demonstration of the training and testing results has shown robustness of the proposed approach against nonlinearities of the experimental datasets and uncertainty in initial SoC. The approach gave satisfactory estimation performance with an acceptable tradeoff between estimation accuracy and convergence speed.

Development of a Hybrid Adaptive Neuro-fuzzy Inference System with Coulomb-Counting State-of-Charge Estimator for Lithium–Sulphur Battery

“…The constant k is the forgetting factor, it allows to gradually discard older data and give more weight to new one [35]. Equation (19) updates the parameters at each step by reducing the error between the predicted output and the measured one. Once the vector ĥðkÞ is calculated, it is 9)-( 12):…”

“…Other battery state estimators that have been studied for Li-S cells use Long Short-Term Memory Recurrent Neural Network (LSTM RNN) [19] or Classification Technique [20]. On the other hand, particle filters and Kalman filterbased SOC estimators have already been used widely in the literature, but applied for Li-ion batteries [21,22].…”

Recent Progress and Emerging Application Areas for Lithium–Sulfur Battery Technology

“…In response to the aforementioned problem, two families of estimation techniques have been proposed for Li-S cells in the literature: 1) techniques derived from control and estimation theory, based on nonlinear variants of the Kalman filter, [82,86] and 2) techniques that come from computer science such as Adaptive Neuro-Fuzzy Inference Systems (ANFIS) [87] and Long Short-Term Memory Recurrent Neural Networks (LSTM RNN). [88] Although most of the research published in the literature are focused on Li-S SoC estimation, the Li-S cell SoH estimation techniques are also under development from both control theory and computer science. Examples are the general framework describing Li-S cell SoH in terms of capacity fade and resistance growth which have been presented in a study by Wild et al [7] and the SoH estimation technique presented in a study by Knap.…”

Online State Estimation for a Physics-Based Lithium-Sulfur Battery Model

“…Existing research on Li-S battery state estimation relies predominantly on either equivalent circuit models or machine learning methods or both. For example, state estimation techniques have been applied to Li-S ECMs in [22,23,24] and to machine learning models in [25]. In [22], the extended Kalman filtering (EKF), unscented Kalman filtering (UKF) and particle filtering techniques are applied and compared for experimental Li-S SOC estimation.…”

“…In [23], an adaptive neuro-fuzzy inference systems algorithm is developed to estimate the SOC based on real-time cell model ECM parameterization. In [24], a dual Kalman filtering technique is used for combined Li-S state and parameter estimation, Finally, in [25], a Long Short-Term Memory Recurrent Neural Network model is built and calibrated for online Li-S state estimation.…”