غير محددة في المصدر 2024 بحاجة إلى مراجعة

An automation system for predictive maintenance of electric machines applied to pumping systems

زيارة المصدر الأصلي

الملخص

This thesis presents the development and implementation of an automation system for predictive maintenance of electric machines in pumping systems. The work integrates Industry 4.0 technologies such as IoT devices, machine learning algorithms, and advanced sensor systems to enhance the reliability and efficiency of industrial operations. A complete test bench was developed, featuring an electric pump, water reservoir, automation and control board, energy meter, vibration sensors, pressure sensors, motorized valves, and temperature sensors, allowing automated test procedures including fault-injection behaviors through valve control. The research bridges the gap between theoretical models of predictive maintenance and their practical implementation in industrial environments, emphasizing the importance of automation and data-driven decision-making. The core achievements include the successful development and deployment of a system capable of real-time data acquisition, advanced vibration analysis, and fault prediction using machine learning. A detailed analysis was conducted to determine optimal data processing procedures prior to analysis, and automation-based sensors were integrated with electronic-based sensors in a unified predictive maintenance system. Machine learning algorithms demonstrated the feasibility of implementing predictive maintenance within standard automation pumping systems by successfully predicting faults induced in the test pump. The developed system not only reduces unexpected failures but also aligns with modern demandsfor sustainability and operational efficiency. By combining data acquisition, real-time analysis, and predictive modeling, the research offers a comprehensive approach that can be adapted across various industries reliant on electric machines.

الكلمات المفتاحية

غير متوفر

التخصصات

رسائل وأطاريح مشابهة