Project title: Immune System Inspired Intelligent Health Monitoring of Rotating Machinery Systems
Sponsors: National Science Foundation, Advanced Diagnosis Automation and Control Lab, North Carolina State University
Duration: Fall 2002 - Spring 2005
Team Members: Bulent Ayhan, Mo-Yuen Chow
Description:

To use Immune System philosophy to dynamically adapt and evolve an Artificial Neural Networks - Fuzzy Logic based Health Monitoring System to explore, learn and absorb new rotating machinery incipient fault detection and discovery knowledge in order to provide accurate health monitoring response in dynamic environments.

The developed Health Monitoring System will be:

  • Self-exploring system based on Immune System philosophy;
  • Less dependent to human involvement;
  • Provide heuristics about the health monitoring processes;
  • Disseminate learned information to other health monitoring systems.
Milestones completed:
  1. MotorSIM2 development;
  2. Collect real time bearing vibration data using Machinery Fault Simulator under different bearing fault conditions and motor speed,
    Analyze frequency spectrum of the collected vibration data for multiple feature extraction;
  3. Investigate, develop, and implement a NN-Fuzzy based structure that can generate heuristics (rules) with respect to the extracted features at different motor speeds;
  4. Define criteria to select the appropriate features in the NN-Fuzzy based structure using Immune System clonal selection principle.

 
Publications:
  • B. Ayhan, M.-Y Chow, H.J. Trussell, M.-H Song, “A Case Study on the Comparison of Non-parametric Spectrum Methods for Broken Rotor Bar Fault Detection,” Proceedings IEEE IECon’03, Roanoke, VA, Nov 2 – Nov 6, 2003.
  • B. Ayhan, M.-Y. Chow, H. J. Trussell, M.-H. Song, E.-S. Kang, and H.-J. Woo, "Statistical Analysis on a Case Study of Load Effect on PSD Technique for Induction Motor Broken Rotor Bar Fault Detection," Proceedings of IEEE SDEMPED ‘03, Atlanta, Georgia, USA, 24-26 August, 2003.
  • M.-H. Song, E.-S. Kang, C.-H. Jeong, M.-Y. Chow, and B. Ayhan, “Mean Absolute Difference Approach for Induction Motor Broken Rotor Bar Fault Detection,” Proceedings of IEEE SDEMPED ‘03, Atlanta, Georgia, USA, 24-26 August, 2003.

 
Bench/Prototype:

Software Prototype - To be added.
Links:

MotorSIM2

Machineary Fault Simulator and Motor Health Moniroting System Benchmarking system