The following code along with the accompanying files, constitutes the primary dataset for the forthcoming research titled "DSP-Based Estimation of Battery State of Charge: Comparative Study on Extended Kalman Filter and Feedforward Deep Neural Networks," which the preprint version has been published at Heliyon First Look
This code is a modified version of the original work by Fauzia Khanum et al. from McMaster University, titled State of Charge Estimation Function based on Kalman Filter
Several modifications have been made to adapt the code to a different battery dataset from the same university, which is available on Mendeley Data - Samsung INR21700 30T 3Ah Li-ion Battery Dataset.
The "Battery Datasets" folder is divided into "Preprocessed Dataset" and "Raw Dataset" subfolders. The Preprocessed Dataset folder contains a list of input data for machine learning training. Please do not modify these files. The Raw Dataset folder contains the original Samsung INR21700 30T 3Ah Li-ion Battery Dataset.
The "Original Code" folder includes the original works by Fauzia Khanum and her colleagues.
Lastly, the "Simulation Results" folder contains a collection of training results and supplementary data for the research paper.