Can the spatial distribution of damping be measured?
Sondipon Adhikari, Jim Woodhouse, A. Srikantha Phani
Abstract
Sondipon Adhikari, Jim Woodhouse, A. Srikantha Phani
Abstract
Our knowledge regarding the detailed description of damping in a vibrating structure is very primitive. The most common approach involves assuming proportional viscous damping, then determining damping factors for each mode. The next step, followed by many authors, is to obtain a full nonproportional viscous damping matrix. Both methods may misrepresent the true damping behavior, as viscous damping is assumed from the outset. In this study a new method for identification of the spatial distribution of damping is proposed. The approach is based on general linear damping models and is not confined to the traditional viscous damping model. Algorithms have been developed for taking measured frequencies, damping factors and complex mode shapes, and fitting the parameters of a damping model with an exponential relaxation function. The viscous damping model is a special case of this model. Simulation studies show that the method seems to perform quite robustly. Preliminary experimental results on a beam with a localized patch of damping are very encouraging and show that the method can indeed indicate the spatial distribution of damping correctly. The approach also gives some information on the question of whether the chosen exponential damping mechanism is correct or not.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Our knowledge regarding the detailed description of damping in a vibrating structure is very primitive. The most common approach involves assuming proportional viscous damping, then determining damping factors for each mode. The next step, followed by many authors, is to obtain a full nonproportional viscous damping matrix. Both methods may misrepresent the true damping behavior, as viscous damping is assumed from the outset. In this study a new method for identification of the spatial distribution of damping is proposed. The approach is based on general linear damping models and is not confined to the traditional viscous damping model. Algorithms have been developed for taking measured frequencies, damping factors and complex mode shapes, and fitting the parameters of a damping model with an exponential relaxation function. The viscous damping model is a special case of this model. Simulation studies show that the method seems to perform quite robustly. Preliminary experimental results on a beam with a localized patch of damping are very encouraging and show that the method can indeed indicate the spatial distribution of damping correctly. The approach also gives some information on the question of whether the chosen exponential damping mechanism is correct or not.
Key concepts: Viscous damping, Damping matrix, Thermoelastic damping, Magnetic damping, Exponential function, Mode (computer interface), Relaxation (psychology), Damping torque