The project focused on "Battery Remaining Useful Life (RUL) Prediction using a Data-Driven Approach with a Hybrid Deep Model combining Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM)." This repository aims to revolutionize battery health estimation by leveraging the power of deep learning to predict the remaining useful life

MATBOX is an open-source MATLAB toolbox dedicated to microstructure analsyis of porous/heterogeneous materials

This paper summarizes a deep learning-based approach with an LSTM trained on the widely used Oxford battery degradation dataset and the help of generative adversarial networks (GANS).

Arduino firmware for a Lithium-Ion battery charger

Electrochemical lithium-ion battery model of reduced-order to predict unmeasurable cell states for fast charging control in real time. Methods for parameterization of the model based on half and full cell measurements are provided.

Deep learning of lithium-ion battery SOH using the DeTransformer model learns the aging characteristics of the battery and then makes predictions about the battery SOH in order to monitor the health of batteries in electric vehicles.

This repository contains code for estimating the State of Charge (SoC) of LG HG2 batteries using Fully Connected Network (FCN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) models along with optuna based hyperparameter tuning.

Data and code for the paper 'Estimating Battery State-of-Charge within 1% using Machine Learning and Physics-based Models' (SAE'23)

Gerbers, BOM, and Schematics for Open UpCell [1S]

A robot car developed using Arduino that can operate in 3 modes - Manual, Automatic and Voice. The car is controlled wirelessly via Bluetooth with an android app developed using MIT App Inventor.

Unofficial reproduction of: A transferable lithium-ion battery remaining useful life prediction method from cycle-consistency of degradation trend(2022)

Grid-scale li-ion battery optimisation for wholesale market arbitrage, using pytorch implementation of dqn, double dueling dqn and a noisy network dqn.

DRL-based Fast Balance Charging Optimization for lithium-ion Batteries

An artificial neural network (ANN) based method is developed for achieving more accurate remaining useful life prediction of Lithium Ion batteries subject to condition monitoring. The ANN model takes the capacity attribute as a target against multiple measurement values as the inputs, and the life expectancy as the output.

Interfacing the laptop battery with the RaspberryPi

This app is an ASE-base workflow used to reproduce a rational initial SEI morphology at the atomic scale by stochastically placing the crystal grains of the inorganic salts formed during the SEI's reaction.

High Dimensional Regression Coefficient Analysis for Functional Data

This repository provides a model deployment framework (MDF) for real-time lithium-ion battery model utilization in CAN-capable test benches. It can be used for the investigation of advanced battery management strategies in short- and long-term experimental studies.

Energy Storage course: practical exercise on the simulation of lithium ion batteries