Data for: Sensitivity-based model update for estimating generalized proportional1 damping parameters in finite element model
Description
These are the files used in the article: Sensitivity method for structural model update and identification in finite elements using vibration measurements. Please, cite and refer to the authors and this reference if you want to make use of the experimental data available here. If you have any question, please contact the authors by email: lamartini_90@hotmail.com. The idea of the experiment was to observe the natural variability that happens in vibration measurement for natural frequencies and damping ratios from multiple measurements taken on different days, after assembling and disassembling the metallic frame. The following files have the objective to help users who want to reproduce the paper's experimental data: 1. "Detailing - Experimental Model.pdf" shows the structure’s drawing and allows one to manufacture the same structure; 2. "CAD.STEP" is the 3D model from structure; 3. "Materials - FE Model.png" shows the materials and Finite Element Model used; 4. "Accelerometers.png" shows the accelerometer locations and hammer impact location; 5. "Figure_1.jpg", "Figure_2.jpg", "Figure_3.jpg" and "Figure_4.jpg" shows the experimental configuration; 6. "ExperimentalData.rar" contains the MatLab code "processamento.m" to post-process all the experimental data. These data are in proprietary Matlab DAQ-files format, readable by daqread command from MatLab. All the data are referred to acceleration time-histories for the 3 accelerometers in the direction perpendicular to the aluminium bars in the positions they were fixed. 6.1. "datastream.daq" file has the acceleration data for each measurement. There are 100 "datastream.daq" files, however, it was used only 50 data streams in that paper (datastream51.daq to datastream100.daq); 6.2."ResultadosExperimentais.xlsx" is a spreadsheet that summarizes the experimental results, mean, standard deviation, and covariance for all 8 output vectors (four natural frequencies and four modal damping ratios); Experimental Setup and equipment: The base of the frame structure was fixed on a rigid table and it was excited by an impact modal hammer. The impact location is presented in the Accelerometers.png file. The ADXL 203 accelerometers, by Analog Devices, were used to measure the vibrations signals from the impact. They have a 970 mV/g sensitivity and a maximum acceleration measurement of ±1.7 g, with a frequency range from 0 to 550 Hz. The acceleration signals were acquired with a Measurement Computing 12-bit USB acquisition board at a sampling rate of 1000 Hz. The same Accelerometers.png file presents the precise accelerometer’s location. The modal hammer is the Endevdo Isotron 4416, with a sensitivity of 1.14 mV/N connected to a signal conditioner Endevco Isotron 4416-1 with 10X gain. The last 50 measurement data were acquired to obtain averages, standard deviations, and covariances of first natural frequencies and the corresponding damping ratios to use in the finite element model updating process.
Files
Steps to reproduce
To reproduce the paper's results it is necessary to follow all the steps below: 1. Download "ExperimentalData.rar" and extract it; 2. Run the "processamento.m" in the matlab; 2.1. Change, in line 3, the value of "kk" to obtain the natural frequencies and modal damping ratio values of this experimental data; 2.2. There are 100 datastream files, representing 100 experimental data, but it was used only 50 datastream in this paper (datastream51.daq to datastream100.daq). So, I must to used the number of datastream (51 to 100) to set the "kk" variable on the line 3 of the code "processamento.m"; 3. Obtain the mean, standard deviation and covariance for all 50 experimental data (8 parameters: 4 natural frequencies and 4 modal damping ratio); The steps above will help users to obtain the experimental results needed to perform the model identification. All the other files attached here (images, CAD file, drawing...) have the objective to give users all the tools to manufacture and execute the same experimental setup. For more details, please see the original paper or contact the authors by email: lamartinifb@gmail.com