Advanced Modeling and Prediction
Explore advanced methodologies for modeling and prediction in the field of battery technologies. These resources focus on leveraging machine learning, physics-based models, and multivariate techniques for accurate predictions and degradation analysis.
-
Data-driven Prediction of Battery Cycle Life Before Capacity Degradation
Description: This dataset supports the prediction of battery cycle life prior to significant capacity degradation. The data enables early detection and modeling for lithium-ion battery lifecycle management.
-
Physics-Informed Neural Network for Battery Degradation Modeling
Description: This dataset includes data for 55 nickel-cobalt-manganese (NCM) 18650 batteries tested under six different charging and discharging strategies. It supports physics-informed neural network modeling for stable degradation analysis.
-
Prognosis of Multivariate Battery State of Health via Transformers
Description: This dataset contains processed files for reproducing results in multivariate battery state prediction using transformers. It includes datasets for lithium-iron-phosphate fast charging and six cathode chemistries.
-
Machine Learning for Degradation Pattern Identification
Description: This dataset supports machine learning-based identification of degradation patterns in lithium-ion batteries through impedance spectroscopy analysis.