Session: 16-01: Poster Presentations

Paper Number: 132021

132021 - Industrial Carbon Footprint Reduction for an Industrial Building via Energy Consumption Optimization Using Machine Learning Tools 

Abstract: 

The increasing population growth and demand for new building construction is known for the most significant factor to greenhouse gas emissions. Therefore, improving the energy efficiency of buildings is one of the most important aspects to reduce greenhouse gas emissions.  Enhancing energy efficiency in buildings stands as a pivotal necessity owing to its profound impact on various crucial facets. Improved energy efficiency not only significantly curtails operational costs but also contributes to a substantial reduction in environmental footprint, playing a pivotal role in combatting climate change. By implementing energy-efficient measures in buildings—ranging from optimized heating, cooling, and lighting systems to advanced insulation and smart technology integration—significant reductions in energy consumption can be achieved, ultimately translating into lower utility bills for building owners. Moreover, the collective effect of energy-efficient buildings contributes to a noticeable decrease in greenhouse gas emissions, industrial carbon footprint, and conserving resources and fostering a sustainable future.

 

This research endeavors to use Artificial Neural Networks (ANN) in enhancing energy efficiency within the Heating, Ventilation, and Air Conditioning (HVAC) systems of corporate environments. The primary aim of this research is to optimize energy consumption in a targeted Industrial building through the application of advanced machine learning techniques. Leveraging ANN, this study seeks to create an intelligent system capable of dynamically adapting HVAC operations to the real-time demands of the building, thereby minimizing energy usage without compromising comfort or functionality. By integrating historical and real-time data from the HVAC systems, the ANN will learn and predict usage patterns, thus allowing for real time operational adjustments and efficient energy management.  

 

In conjunction with the ANN implementation, this research project will employ EnergyPlus software to model and analyze energy consumption within the company building. The integration of EnergyPlus will serve as a valuable complementary approach to validate and fine-tune the proposed energy optimization strategies. By utilizing EnergyPlus, a comprehensive and detailed simulation of the company's energy usage patterns will be developed. This simulation will serve as a benchmark for the performance evaluation of the ANN-driven energy optimization system within the HVAC infrastructure. The combined approach of ANN and EnergyPlus software will provide a holistic and reliable framework for devising, testing, and implementing energy-efficient protocols in corporate settings. The aim of this research in not only to reduce the company energy consumption and industrial carbon footprint, but also to contribute to the establishment of sustainable and environmentally energy management practices in the company.

Presenting Author: Mohammad Hadi Katooli Purdue University, Indianapolis

Presenting Author Biography: Mohammad Hadi Katooli is a Ph.D. student at Purdue University, Indianapolis. He has B.Sc mechanical engineering from Shahid Beheshti university and MS. in mechanical engineering in University of Tehran. He is currently a research assistant and a DOE-certified energy auditor at Industrial Assessment Center (IAC) lab, Purdue University, Indianapolis. He has published over ten papers in international journals. His research intertest is optimization of energy consumption using machine learning.

Authors:

Mohammad Hadi Katooli Purdue University, Indianapolis
Ali Razban Purdue University, Indinapolis