Using model order reduction techniques for fault detection system in METIS Cold Chopper

Alkano, D. (2011) Using model order reduction techniques for fault detection system in METIS Cold Chopper. Master's Thesis / Essay, Industrial Engineering and Management.

	Text 2011_Alkano_Destie_MSc-DT.pdf - Published Version Restricted to RUG campus only Download (2MB)
	Text AkkoordDAlkano.pdf - Other Restricted to Registered users only Download (51kB)

Abstract

The Mid-Infrared Imager and Spectrograph (METIS) is one of important science instrument being proposed for the next generation of European Extremely Large Telescope (E-ELT). The METIS cold chopper (MCC) has been identified as one of the critical components in the whole development of METIS. This research examines this critical component. A material deformation may occur on the MCC mirror once METIS is on schedule. In this respect, the critical failure can disrupt the normal operation of the telescope and affect the scheduling of observation. At that time, high cost due to inaccurate data is unavoidable and the maintenance requires the whole process of observation to be stopped for repairing. Regarding the unexpected event mentioned above, a continuous monitoring of the MCC is becoming essential. By providing online monitoring, the MCC system is expected to be able to prevent the unexpected event. Early warnings on possible fault may help the maintenance department to plan an appropriate maintenance action. By making use of the early warnings, the timely procurement of the spare parts and planning of the replacement are expected to be well prepared. Dynamical model of MCC mirror structure is needed in the design of the fault monitoring system. Timoshenko beam theory is used to develop a simplified model of MCC mirror. The finite element methods (FEMs) have been widely used in the analysis of the dynamics of the mechanical systems. However, the FEM model is computationally expensive to be implemented in a fault detection system of MCC. Using model order reduction (MOR) techniques, a reduced-order model can be obtained from the high-order FEM model and, subsequently, be used in the low-order fault detection system. Two MOR techniques are investigated: balanced truncation method and moment matching method. The simulation results show that the balanced truncation-based fault detection is lower computational load, better in approximating the transfer function of the high-order FEM model, and better classifier than the moment matching-based fault detection. The quality of balanced truncation and moment matching depend on the number of order preserved in the reduced model and the chosen expansion

Item Type:	Thesis (Master's Thesis / Essay)
Degree programme:	Industrial Engineering and Management
Thesis type:	Master's Thesis / Essay
Language:	English
Date Deposited:	15 Feb 2018 07:46
Last Modified:	15 Feb 2018 07:46
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/9804

Actions (login required)

View Item