Abstract:Predicting the evolution of multiphysics battery systems face severe challenges, including various aging mechanisms, cell-to-cell variation and dynamic operating conditions. Despite significant progress, solving real-life battery problems with noisy and missing data and high-dimensional parameter space are either difficult or impossible. In this paper, we design and evaluate feature-based machine learning techniques for estimating the capacity of large format LiFePO4 batteries in EV applications and hence predicting the trajectory of capacity fade based on the estimations. To probe the feature space, we generate a comprehensive dataset consisting of 420 cells and 9 battery packs (178 cells in-series for each one) with more than 10,000 validation data derived from the cloud platform. A two-step noise reduction method is applied to de-noise the scattered field data (30s sampling interval). Totally, 39 domain features are engineered using the reconstructed segments of battery charging data based on the differential methods (increment capacity and differential voltage), which steer the learning process towards accurate and physically consistent predictions by leveraging the stacking ensemble learning. The stacking, comprised of four base learning models referred to as level-1 predictors and a meta-learner referred to as level-2 predictor applied to combine predictions of base learners with probability distributions using an extended set of meta-level features offers exciting opportunities for better accuracy and improved generalization. Our best models achieve 0.28% mean absolute percentage error (MAPE) and 0.55% root mean squared percent error (RMSPE) for battery capacity estimation. Further, 1.22% average percentage error is achieved in the prediction of remaining useful lifetime (RUL) under different conditions of driving distance (km) and service time (day) by building capacity fade trajectory based on a Bayesian regression. This work highlights the promise of machine learning modelling using domain-specific features for accurate estimation and prediction of real-life battery systems based on field data collection.

Unsupervised feature extraction for lithium-ion battery electrochemical impedance spectroscopy and capacity estimation using deep learning method

A Physics-Informed Hybrid Data-Driven Approach with Generative Electrode-Level Features for Lithium-Ion Battery Health Prognostics

Impedance-based capacity estimation for lithium-ion batteries using generative adversarial network

Capacity estimation of lithium-ion batteries with automatic feature extraction and graph-enhanced LSTM

Automatic Feature Extraction Enabled Lithium-Ion Battery Capacity Estimation Using Random Fragmented Charging Data

Battery pack capacity estimation for electric vehicles based on enhanced machine learning and field data

A transfer learning-based ensemble learning model for electric vehicles lithium-ion battery capacity estimation using electrochemical impedance spectroscopy

Semi-supervised estimation of capacity degradation for lithium ion batteries with electrochemical impedance spectroscopy

A Hybrid Drive Method for Capacity Prediction of Lithium-Ion Batteries

Transferable data-driven capacity estimation for lithium-ion batteries with deep learning: A case study from laboratory to field applications

Lithium-ion battery health prognosis via electrochemical impedance spectroscopy using CNN-BiLSTM model

Machine learning for predicting battery capacity for electric vehicles

A Data Augmentation Method for Lithium-Ion Battery Capacity Estimation Based on Wassertein Time Generative Adversarial Network

A Hybrid Convolutional Neural Network-Long Short Term Memory for Discharge Capacity Estimation of Lithium-Ion Batteries

An Enhanced Data-Driven Model for Lithium-Ion Battery State-of-Health Estimation with Optimized Features and Prior Knowledge

Convolutional autoencoder-based SOH estimation of lithium-ion batteries using electrochemical impedance spectroscopy

State of Health estimation method for lithium batteries based on electrochemical impedance spectroscopy and pseudo-image feature extraction

A novel variable activation function-long short-term memory neural network for high-precision lithium-ion battery capacity estimation

A novel approach for prognosis of lithium-ion battery based on geometrical features and data-driven model

State of health and remaining useful life prediction for lithium-ion batteries based on differential thermal voltammetry and a long and short memory neural network