The project analyzes battery cycling data to predict degradation patterns and performance metrics using both deep learning (LSTM) and traditional machine learning (XGBoost) approaches. The implementation enables accurate estimation of battery health, which is crucial for battery management systems in various applications.

Neural ODE-based State of Charge (SOC) estimation for Li-ion batteries using the NASA Battery Dataset. Built as a weekend project to explore learned dynamics for battery modeling, with visualizations designed for engineering audiences.

This repository contains code for analyzing battery data from NASA's battery testing dataset. The analysis involves processing battery impedance, electrolyte resistance, and charge transfer resistance across charge/discharge cycles to track the aging and performance of various batteries.

Implementácia metód Isolation Forest, One-Class SVM a Autoencoder na detekciu anomálií pri nedostatku anomálnych dát na datasetoch NASA Battery, NAB a ECG5000.

NASA battery RUL prediction workflow with feature extraction, analysis notebooks, and ML regression benchmarks.

Implementation of Isolation Forest, One-Class SVM, and Autoencoder Methods for Anomaly Detection under Limited Anomalous Data Using the NASA Battery, NAB, and ECG5000 Datasets.