IEEE Transactions on Industrial Electronics · 2020 · 785 citations · 30 references
ComputingComputational EngineeringData ScienceMachine LearningPredictive AnalyticsUncertainty QuantificationBattery DiagnosticsBattery SystemsBattery TechnologyCapacity ForecastingReliability ModelingUncertainty AnalysisBattery Aging ModelsBattery ModelingBattery PrognosticsEmpirical Mode DecompositionGaussian Process RegressionLong Short-term Memory NetworksRemaining Useful Life PredictionUncertainty Management
Predicting future capacities and remaining useful life (RUL) with uncertainty quantification is a key but challenging issue in battery health diagnosis and management. The study applies advanced machine‑learning techniques to predict future capacities and RUL of Li‑ion batteries with reliable uncertainty management. The authors decompose capacity data using empirical mode decomposition, then model the residual with an LSTM and the intrinsic mode functions with a Gaussian process regression, jointly capturing long‑term dependence and uncertainty, and compare the combined LSTM+GPR model against several baselines on experimental aging data. The combined LSTM+GPR model outperforms other methods, delivering accurate 1‑step and multistep capacity predictions and reliable uncertainty‑quantified RUL estimates even at early battery cycles.
Predicting future capacities and remaining useful life (RUL) with uncertainty quantification is a key but challenging issue in the applications of battery health diagnosis and management. This article applies advanced machine-learning techniques to achieve effective future capacities and RUL prediction for lithium-ion (Li-ion) batteries with reliable uncertainty management. To be specific, after using the empirical mode decomposition (EMD) method, the original battery capacity data is decomposed into some intrinsic mode functions (IMFs) and a residual. Then, the long short-term memory (LSTM) submodel is applied to estimate the residual while the Gaussian process regression (GPR) submodel is utilized to fit the IMFs with the uncertainty level. Consequently, both the long-term dependence of capacity and uncertainty quantification caused by the capacity regenerations can be captured directly and simultaneously. Experimental aging data from different batteries are deployed to evaluate the performance of proposed LSTM+GPR model in comparison with the solo GPR, solo LSTM, GPR+EMD, and LSTM+EMD models. Illustrative results demonstrate the combined LSTM+GPR model outperforms other counterparts and is capable of achieving accurate results for both 1-step and multistep ahead capacity predictions. Even predicting the RUL at the early battery cycle stage, the proposed data-driven approach still presents good adaptability and reliable uncertainty quantification for battery health diagnosis.
30
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
Jun‐Young Chung, Çağlar Gülçehre, Kyunghyun Cho et al. · arXiv (Cornell University) · 2014 · 10.8K citations · Full text