Predictive Models for Bioenergy in the Agroindustrial Sector of São Paulo Using Machine Intelligence

• Grant number: 24/19995-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: February 01, 2025
• End date: January 31, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Marilaine Colnago
• Grantee: Lívia Gabriela Souto de Souza
• Host Institution: Instituto de Química (IQ). Universidade Estadual Paulista (UNESP). Campus de Araraquara. Araraquara , SP, Brazil

Abstract

The present project aims to develop advanced computational models to estimate bioenergy production from sugarcane in the state of São Paulo, using real data from various sources. The methodologies employed include Random Forest Regression (RFR), Support Vector Regression (SVR), and Deep Neural Networks (DNN). Initially, public datasets provided by different institutions will be explored visually and analytically, with a focus on data preprocessing and adaptation for predictive models. In addition to implementing, training, tuning, and validating the models based on at least five distinct datasets, an Exploratory Data Analysis (EDA) will be conducted to maximize data understanding and create a robust information repository, which can be shared with managers in São Paulo's energy sector. The expectation is that the developed models will provide accurate predictions of bioenergy production, supporting the optimization of operations and strategic decision-making in São Paulo's agro-industrial sector.