Experimental data: Intruder dynamics in granular media under localized surface loading
Description
This dataset contains experimental data of an intruder moving in a granular medium whose surface is loaded with an extra weight. In the experiments, described in the paper "Intruder dynamics in granular media under localized surface loading", a 12-mm-diameter intruder was forced to move under constant force within 1-mm-diameter glass particles (granular medium), and an extra weight was placed on part of the the surface of the granular medium. We varied the weight of the extra load and the depth of the intruder, and found that a deceleration occurs when the intruder passes under the extra weight. The dataset corresponds to txt files (in the CSV format, separated by space) with the time (in seconds) and position (in mm), that were obtained from image processing of high-speed movies (not included here given their large size). The dataset also contains an index file named index_mass.txt, which contains the filename (of each of other txt files), the mass of the extra load (in g), the depth of the intruder (in mm), and the area of application of the extra load (in mm2). From these files, it is possible to compute everything that is shown in the mentioned paper. The dataset contains also phyton codes written to process the txt files. We observe that the intruder moves in the quasistatic regime, and that when an extra load is added over the granular surface the intruder decelerates in the region beneath it. The effect is observed even at the highest depth tested, varying with both the depth and the weight of the extra load.
Files
Steps to reproduce
1) Download and decompress the file tests_all.zip. 2) Download the file index_mass.txt. 3) Open the code postproc_new.py and include the correct path to the folder where you placed the txt files. 4) Select the options in the INPUT portion of the postproc_new.py code (in its beginning) and run the code. It is important to set 'save_data = True' in order to use the other codes (described in the next steps). The postproc_new.py code will read the index_mass.txt and each file of the test runs, will organize them in dataframes, and, for each test run, will find the beginning and end of motion, compute velocities and accelerations, smooth the data (by moving averages), and computing pertinent dimensionless numbers and characteristic times. After all that, it will save output as CSV files. 5) Run the postproc_post.py code to open an CSV file generated in step 4, compute new characteristic times, fit the deceleration regions of overloaded tests, and find the effective overpressure acting on the intruder (but note that it depends on the values of Kz chosen). The code will print the main results and plot the results used in the referred paper (some of the plot commands are commented, just un-comment them for having the plot on your screen). It is important to have set 'save_data = True' in postproc_new.py. 6) This step is not necessary, unless you wish to fit the velocities of tests without overload with the model developed in the paper, and obtain the values of Kz for those tests. For that, simply run fitting_velocity.py, and the code will print characteristic quantities and values of Kz, besides plotting the results. It is important to have set 'save_data = True' in postproc_new.py.
Institutions
- Université Paris-SaclayÎle-de-France, Gif-sur-Yvette
- Universidade Estadual de CampinasSP, Campinas