WLAN-enabled Sensor Nodes for Cloud-based Machine Condition Monitoring☆

Abstract In the recent past, Machine Condition Monitoring (MCM) has gained popularity, in order to extend equipments’ lifetime and reduce unscheduled downtime. The original contribution of this work is the proposal of a WLAN-enabled sensor node capable to locally store prolonged acquisition sessions. Log chunks are periodically sent directly to the Internet, so that MCM-meaningful data can be analyzed by means of cloud services. Traditionally, MCM is based on long-term complex analysis of costly, time consuming and manually-executed measurements of vibrations, thermal profiles and other significant quantities. Both wireless sensor networks and cloud services have been already proposed to overcome these limitations. However, such an approach usually requires one or more gateways for sensor data to converge towards IP-based infrastructure. This network architecture is overcome by the proposed approach, as verified by means of a proof-of-concept prototype, highlighting pros and cons and effectiveness of the solution.

[1]  G. Marichal,et al.  An intelligent system for faulty-bearing detection based on vibration spectra , 2011 .

[2]  Yuan Feng,et al.  CCSA: A Cloud Computing Service Architecture for Sensor Networks , 2012, 2012 International Conference on Cloud and Service Computing.

[3]  Emiliano Sisinni,et al.  Performance assessment of vibration sensing using smartdevices , 2014, 2014 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings.

[4]  Michael J. Devaney,et al.  Adjustable Speed Drive Bearing Fault Detection via Wavelet Packet Decomposition , 2006, 2006 IEEE Instrumentation and Measurement Technology Conference Proceedings.