Input-output vibration-response data of a linear guide considering local and global experimental setups

Input-output vibration-response data of a linear guide considering local and global experimental setups Jens Kristian Mikkelsen Gudichsen, Luis David Avendaño-Valencia Description This dataset accompanies our journal publication for Data-Centric Engineering: Gudichsen, Jens Kristian Mikkelsen and Schlette, Christian and Avendaño-Valencia, Luis David, ’A generalized multi-output identification framework for configuration-dependent structures considering local and global experimental data’, Data-Centric Engineering (2025) The dataset contains the vibration response and excitation signal based in measurements from a linear guide comprised by a cart supported on two parallel flexible rails. The cart position is controlled by a belt system driven by a stepper motor installed on one of the ends of the guide. The supports of the linear guide have been modified so that the rails seat in a fixed-fixed configuration. An electrodynamic shaker with a small mass attached to the stinger is mounted on top of the cart to act as a controllable dynamic load. The cart horizontal dimension is 300 mm and can move 1500 mm on the horizontal axis of the total 1800 mm long linear guide. The setup and the structure are seen in Figure 1: Figure 1: Linear guide system: (a) Linear guide setup schematic and sensor location. Red markers 1-8 are the linear guide output response accelerometers, orange markers 9-10 are the cart output response accelerometers, blue markers 11-12 are the shaker and motor input response accelerometers, respectively. (b) Physical system with sensors and peripherals indicated. (c) Detail of the cart with mounted shaker. Figure 1 and Table 1 indicate the position of the sensors along the beam and the shaker as well as their respective number. Eight accelerometers are installed in pairs along the length of the linear guide rails (indicated with numbers 1-8), two more are installed symmetrically over the cart (indicated with numbers 910), one accelerometer is installed over the mass on the shaker, and the last one is located over the stepper motor. Sensor No. Location (X,Y) [mm] Orientation Type 1 (300,0) Z+ Output 2 (300,250) Z+ Output 3 (660,0) Z+ Output 4 (660,250) Z+ Output 5 (1020,0) Z+ Output 6 (1020,250) Z+ Output 7 (1500,0) Z+ Output 8 (1500,250) Z+ Output 9 Cart Position + (0,0) Z+ Output 10 Cart Position + (0,250) Z+ Output 11 Cart Position + (0,125) Z+ Input 12 (1800,100) Z+ Motor reference Table 1: Summary of sensor location, orientation and type The stepper motor accelerometer signal is used as a reference to estimate the position of the cart over the beam. A white noise generator (100 kHz bandwidth) and power amplifier are used to provide the excitation signal for the shaker. All the 12 accelerometer response signals are fed to a National Instruments data acquisition system comprised by 3 NI-9234 modules fixed to a DAQ-9174 chassis. The remaining components for instrumentation are seen in Table 2: Component Description Accelerometers 12x Dytran 3079A2T DAQ Chassis National Instruments compact-DAQ cDAQ-9174 DAQ Modules 3x National Instruments NI-9234 Power Amplifier Brüel & Kjær Type 2706 Noise Generator Brüel & Kjær Type 1405 Shaker LDS V201-1 Motor Drive Controller Igus drylin® D1 Linear Guide Igus ZLW-20200S-7TH-IO-01-1500 (7th axis) Load 100g load for shaker mounting Table 2: Summary of the instrumentation details used in the linear guide experiments. The main aim of the dataset is to be used to identify the dynamics of the configuration-dependent structure. To this end, local and global vibration tests are performed. Local tests involve the system's vibration response with the cart located at discrete positions over the linear guide, while being excited by an electrodynamic shaker installed on top of the cart. Global tests involve the system's vibration response as the cart moves back and forth over the linear guide while the shaker is also active. In the latter case, the shaker and stepper motor act as excitations to the system. Data is collected at a sampling rate of 5120 Hz for a period of 300 s both for the local and global identification experiments. The sampling rate is selected much higher than the resonance frequencies of the linear guide, to facilitate the separation of the stepper motor excitation from the structural dynamics of the linear guide. Contents The dataset contains 2 zipped data folders with accompanying scripts to present most of the raw data. This is organised as follows: • Three repeated local experiments in the folder “Static experiments” with each local dataset in the subfolders “ShakerSetup1”, “ShakerSetup2” and “ShakerSetup3”. o Each subfolder has 16 files, named “PosXXX.mat”. These are the measured data from all 12 sensors. XXX indicates the static position of the cart in centimetres (e.g. 130 cm or 010 cm). • Three repeated global experiments in the folder “Dynamic experiments” with each dataset named “DynSweep1.mat”, “DynSweep2.mat” and “DynSweep3.mat” which is the measured data from all 12 sensors. • Scripts to load data and perform preprocessing. o “test00_SampleSignals.m” loads the data from both a static and a global experiment and compares Power Spectral Density estimates of the response. o “test01_SpectrogramAnalysis.m” loads the data from the global experiment and downsamples the signal and use a Chebyshev filter to avoid aliasing before the downsampling. Next, the script performs spectrogram analysis of the global experiments. Structure of local experiments files The local experiments files are sorted by cart position on the beam, “PosXXX.mat”. The files contain the vibration response and excitation directly extracted by the National Instruments compact-DAQ. Each individual file consists of 12 vectors of the sampled acceleration data from each sensor, in sequential order from the definitions of the naming scheme in Figure 1. Structure of global experiments files The global experiments files have three files where the cart moves across the beam in a continuous fashion and are named “DynSweepX.mat”. The files contain the vibration response and excitation directly extracted by the National Instruments compact-DAQ, just like for the static experiments. Each individual file consists of 12 vectors of the sampled acceleration data from each sensor, in sequential order from the definitions of the naming scheme in Figure 1. Using and citing this dataset This dataset is distributed under the Creative Commins Attribution 4.0 International license. Further information can be found on: https://creativecommons.org/licenses/by/4.0/ . When using this dataset, please cite the Zenodo link under DOI: 10.5281/zenodo.17475143 Contact Information Jens Kristian Mikkelsen Gudichsen [email protected] SDU Center for Large Structure Production, The Maersk Mc-Kinney Moller Institute University of Southern Denmark Campusvej 55, 5230 Odense M, Denmark