Sensor Networks for Condition Monitoring of Industrial Assets


ECTS Credits

  • prof. dr. Đani Juričić
  • None


Course objectives: - understanding the taxonomy of sensor networks in the area of condition monitoring, - becoming familiar with basic concepts of condition monitoring, - getting acquainted with basic signal processing algorithms in sensor networks, - making students familiar with practical problems. Competences: - ability to perform requirements analysis for condition monitoring systems, - ability to configure a sensor network for system diagnostics and prognostics, - ability to design algorithms for sensor fusion, - ability to make use of MIMOSA database.


- The role of condition monitoring in the maintenance of industrial assets. - The taxonomy of asset condition monitoring and prognostics of the remaining useful life. - Components of the industrial sensor networks, sensors and micro-sensors, smart nodes, systems on chip, gateways and protocol converters. - Feature extraction, Fourier and wavelet transforms, high-order spectra, spectral kurtosis, entropy indices. - Model based approaches: residuals, (non-linear) Kalman filter. - Trend change detection, feature fusion. - Prognostics: statistical approaches based on historical records. - Data acqusition, storage and display in sensor networks, integration with existing information systems, MIMOSA OSA EAI standard. - Industrial case studies. - Individual study of a real case from student’s research work.


Completed second cycle studies in natural sciences or engineering or completed second cycle studies in other fields with proven knowledge of fundamentals in the field of this course (certificates, interview).


Literature and references