Concept and development of IoT-based e-maintenance platform for demonstrated system

An e-maintenance system is critical emerged in modern manufacturing linked to the era of industry 4.0. Such an intelligent system can process real-time monitoring on the whole production line for diagnosing and preventing potential failures. It requires various IoT sensors and modern technological devices as well as a smart algorithm for processing essential signals. Machine learning (ML) has currently been included for smart predictive maintenance. This leads the possible and flexible solutions for intelligent fault diagnosis and prognosis systems. In this paper, the conceptual algorithm of IoT-based e-maintenance has been developed implementing on demonstration of the intelligent manufacturing system as so-called “Demo Site”. The simple and cost-effective system is purposed by using the Arduino UNO board for data collection and analysis of the essential signal. The data is then transferred and real-time displayed on Blynk’s dashboard via the ESP32 connected to the Wi-Fi router. Particular sensors for 6 various working functions are attached to the suitable areas on the demo site system. The data acquisition and signal monitoring algorithm are conducted by Arduino IDE which is C-based standard coding. Thus, spindle speed, spindle vibration, tool holder temperature, coolant level, temperature inside the control cabinet, and conveyor speed are detected and monitored. This allows collecting and analysing machine system’s performance for predictive maintenance decisions in any time from elsewhere. The remarkable results show that the e-maintenance system is capable to analyse demo site system’s errors and deteriorate rate when collected and analysed data is sufficient. Predictive maintenance is thus feasible if information data of machinery is continuously and sufficiently collected and analysed through the use of IoT-based e-maintenance systems and machine learning algorithms.