TY - JOUR
T1 - Modeling energy efficiency in industrial plants
T2 - A novel diagnostic approach
AU - Vásquez Capacho, John William
AU - Perez-Zuñiga, Carlos Gustavo
AU - Ospino-Castro, Adalberto
N1 - Publisher Copyright:
© 2024
PY - 2025/2/15
Y1 - 2025/2/15
N2 - In industrial plants, diagnosing energy efficiency issues is essential to achieve sustainable operations and reduce costs. This paper introduces a novel diagnostic approach using advanced modeling techniques to identify inefficiencies in energy consumption within industrial environments. The proposed method uses discrete event analysis to detect and characterize abnormal energy usage patterns, providing a systematic framework for diagnosing performance issues in complex systems. Two case studies involving high-performance computing (HPC) systems illustrate the practical application of the approach, showcasing its ability to uncover critical inefficiencies and inform energy management strategies. The research addresses a significant gap in current methodologies by providing a detailed diagnostic tool customized to the unique challenges of industrial energy management. This study paves the way for future research into advanced diagnostic techniques, strengthening the importance of precise and actionable information on energy use for industrial stakeholders.
AB - In industrial plants, diagnosing energy efficiency issues is essential to achieve sustainable operations and reduce costs. This paper introduces a novel diagnostic approach using advanced modeling techniques to identify inefficiencies in energy consumption within industrial environments. The proposed method uses discrete event analysis to detect and characterize abnormal energy usage patterns, providing a systematic framework for diagnosing performance issues in complex systems. Two case studies involving high-performance computing (HPC) systems illustrate the practical application of the approach, showcasing its ability to uncover critical inefficiencies and inform energy management strategies. The research addresses a significant gap in current methodologies by providing a detailed diagnostic tool customized to the unique challenges of industrial energy management. This study paves the way for future research into advanced diagnostic techniques, strengthening the importance of precise and actionable information on energy use for industrial stakeholders.
KW - Artificial intelligence
KW - Diagnosis
KW - Discrete-time systems
KW - Energy efficiency
KW - Industry 4.0
UR - http://www.scopus.com/inward/record.url?scp=85213024594&partnerID=8YFLogxK
U2 - 10.1016/j.engappai.2024.109777
DO - 10.1016/j.engappai.2024.109777
M3 - Article
AN - SCOPUS:85213024594
SN - 0952-1976
VL - 142
JO - Engineering Applications of Artificial Intelligence
JF - Engineering Applications of Artificial Intelligence
M1 - 109777
ER -