An Architecture for Distributed Search and Data-Mining in Condition Monitoring Applications

There is an increasing growth in the volume of data generated by condition health monitoring applications as the technology becomes more pervasive and as the sensing technology becomes more complex. This can lead to significant problems in processing the volumes of data in an efficient way, particularly when the data is held remotely. This paper describes a distributed grid architecture that supports real-time pattern matching analysis to address this requirement within complex CHM problems. The architecture is generic and scalable.

[1]  Thomas Jackson,et al.  Predictive Maintenance , 2004, The Grid 2, 2nd Edition.

[2]  Jim Austin,et al.  DAME: Searching Large Data Sets Within a Grid-Enabled Engineering Application , 2005, Proceedings of the IEEE.

[3]  Victoria J. Hodge,et al.  Scalability of a distributed neural information retrieval system , 2002, Proceedings 11th IEEE International Symposium on High Performance Distributed Computing.

[4]  Victoria J. Hodge,et al.  An evaluation of standard retrieval algorithms and a weightless neural approach , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[5]  Lionel Tarassenko,et al.  A System for the Analysis of Jet Engine Vibration Data , 1999, Integr. Comput. Aided Eng..

[6]  Jim Austin,et al.  Chemical similarity searching using a neural graph matcher , 2005, ESANN.

[7]  Jim Austin,et al.  Chemical structure matching using correlation matrix memories , 1999 .

[8]  Allan Gottlieb,et al.  Highly parallel computing , 1989, Benjamin/Cummings Series in computer science and engineering.