Simulation of Experimental Modal Analysis

OData support
Dr. Fiala Péter
Department of Networked Systems and Services

Our environment is full of mechanical vibration systems, think about built landmarks, mechanical creations, musical instruments and daily used instruments. The description, study and measurement of their behavior is one of the most significant engineering tasks. That is why there are loads of developed methods of modelling, counting, measuring and signal processing to determine the most important vibrational parameters. The summary name of the topic is modal analysis.

In my thesis, the target was the implementation and testing of an up-to-date modal analysing algorithm operating in time domain. The algorithm specifies the modal parameters from measured responses of a vibration system. I implemented this in Matlab, and then for testing I created some different models, at first a general mass-stiffness-damping concentrated-parameter model, after that an ideal, homogenous clamped stringmodel as a distributed-parameter model. Based on the created models I specified impulse responses of different testing cases, and then I computed the modal parameters in analytic way. After this I tested the algorithm with the impulse responses as input parameters, and I compared the algorithmic results to the analytic results, in all the test cases, then I also paid attention to the examination of the noise sensitivity of the algorithm, I realized this with the added white-noise of the impulse responses, with taking in account different signal-to-noise ratios.


Please sign in to download the files of this thesis.