Modernisation of fault detection for diagnosis routines in elevators

Abstract

Abstract Maintenance of elevators has become critical in ensuring continued operation by preventing excessive wear and breakdown. Maintenance greatly affects elevator downtime and uptime, hence the need for modernising elevator maintenance to stay abreast of other competitors. This research focuses on the modernisation of maintenance in elevator systems to reduce breakdowns through scheduled maintenance via remote condition monitoring for fault detection using the Internet of Things (IoT) technology. The monthly scheduled maintenance policy for the elevator system, however, increased the downtime of the system due to lengthy response time to attend to elevator breakdowns. This research therefore adopts remote monitoring of the elevator system’s condition for early detection of malfunctioning and faults notification for a just-in-time maintenance response. The parameters which could indicate a fault, deterioration, or damage of the elevator system were identified. The methodology embraced building and configuring an electronic monitoring device which comprises of the sensors, LED light, a voltage source, breadboard, jumper wires and an IoT microcontroller. The microcontroller is programmed to monitor temperature, 3 axial vibration, and acoustics parameters of the elevator system. Data and fault notifications are sent to a registered email for remote monitoring access on the cloud. The IoT devices and controller make use of any back up system which can be accessed in the cloud as a secondary storage system for the data being read by the sensors and notification updates. The back-up system used in this research is electronic mail. The read data from the machine was posted, together with the fault notification in cases of malfunctioning of the condition, to an email cloud server. The results show that remote condition monitoring of the elevator system is a better maintenance approach as it reduces the downtime of the elevator system through just-in-time fault notification, trend monitoring for fault troubleshooting and also diagnosis of fault from historical events. This is indicated by a considerable reduced response time, (81%) as compared to the initial state of the system, with a total response time of 45.4 hours for the 6 fault notifications experienced during the condition monitoring unlike 240 hours for 4 breakdowns before modernising the maintenance approach. Five of the six breakdowns experience were indicated by both vibration and acoustics parameters which shows they are complimentary in fault diagnosis. An optimised limit for each parameter was also derived using control chart for variables analysis.