نوع مقاله : مقاله پژوهشی
نویسندگان
1 کارشناسی ارشد، گروه سیستم، دانشکده مهندسی صنایع، دانشگاه خواجه نصیرالدین طوسی، تهران، ایران.
2 استادیار، گروه مهندسی نگهداری و تعمیرات، دانشکده علوم و مهندسی، دانشگاه افسری و تربیت پاسداری امام حسین(ع)، تهران، ایران.
3 مدرس، گروه مهندسی نگهداری و تعمیرات، دانشکده علوم و مهندسی دفاعی، دانشگاه افسری و تربیت پاسداری امام حسین (ع)، تهران، ایران.
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Introduction: In today's world, preventive maintenance of building facilities, particularly electric motors, is of paramount importance due to their critical role in water supply, heating, and cooling systems. With the advent of Industry 4.0 technologies, utilizing data mining and machine learning for failure prediction and optimizing preventive maintenance has become a necessity.
Objective: This study aims to develop a predictive model for the preventive maintenance of electric motors in building facilities using machine learning and data mining algorithms.
Methodology: This applied-developmental research employs a mixed-methods approach (qualitative-quantitative). The data consists of two main parts: 1) Records from the maintenance system (137) of Military Training Center, comprising 6,848 repair records, and 2) Simulated operational data for an electric motor, including variables such as temperature, rotational speed, torque, and tool wear. After preprocessing the data using Python, various machine learning algorithms, including Random Forest, Decision Tree, Support Vector Machine (SVM), and Logistic Regression, were evaluated.
Findings: The Random Forest algorithm demonstrated the best performance with an accuracy of 99.6% on the test data. This model can predict electric motor failures with high accuracy and identify hidden patterns in operational data. The results also indicate that this approach can reduce unexpected downtime, extend equipment lifespan, and optimize maintenance costs.
Conclusion: This research demonstrates that integrating data mining and machine learning can be effectively applied for the predictive maintenance of electric motors in building facilities, leading to improved system reliability, safety, and operational efficiency.
کلیدواژهها [English]