Etudiant : ZEROUAL SOUKAYNA

Filière : Master Systèmes Intelligents et Décisionnels (MSID)

Encadrant : Pr. AHERRAHROU NOURA

Annèe : 2023

Résumé : This report serves as a comprehensive summary of my end-of-study internship project conducted at MAScIR. During this internship, I had the opportunity to work on a project focused on optimizing grinder energy consumption in the mining industry. Grinders play a significant role in energy usage during the mining process, and improving their efficiency is crucial for reducing operational costs and minimizing environmental impacts. This internship was conducted in collaboration with Managem Group and Reminex SA, as part of the Moroccan national project ”Smart Connected Mine”. The project brings together industrial stakeholders and researchers to drive the digital transformation of the mining industry. Working collaboratively with Managem Group and Reminex SA, valuable data from the mining industry was made accessible. The data underwent meticulous preprocessing, where advanced feature selection and engineering techniques were employed to extract significant insights from the dataset. Throughout the internship, my primary objective was to develop a regression model using machine learning and deep learning techniques, specifically of Long Short-Term Memory (LSTM) and Artificial Neural Network (ANN) techniques. Additionally, I employed a Genetic Algorithm (GA) optimization algorithm to identify the best parameters for the grinders. The ultimate goal was to prevent over grinding and under grinding, promoting more efficient and sustainable grinding practices while minimizing operational costs and environmental impacts.