Acoustic Emission dataset for impact localization: numerical and experimental case studies 机器收录·待核验
mxdsj4g8j2e24y07
Acoustic Emission dataset for Defect Detection in Aluminum plates
Simulated data
File name: Simulation.zip
Simulated AE signals based on a ray-tracing algorithm taking into consideration reflections with the mechanical boundaries of the medium (reflection up to the 4th order), which corresponds to a 1x1x0.003 m square aluminum plate.
Each signal has been created by simulating the propagation between a transmitter (Tx, index from 1 to 40) and a Receiver (Rx, index from 1 to 25), grouped by Tx position and saved as a .mat struct ('data') containing the following fields:
data.Rx = 2 x 25 matrix containing in the first and second row the x and y axis of the Rx position
data.Rx = 2 x 25 matrix containing in the first and second row the x and y axis of the Tx position
data.data = 8000 x 25 matrix containing the transmitted, propagated AE signal from Tx to Rx organized by column. Each AE instance constitutes of 8000 samples acquired at a sampling frequency of 2 MHz (indicated in the file name), one for each Rx given that Tx position.
data.Label = 1x25 vector containing the ToA labels associated with each of the 25 Tx-rx pairs (per Tx position) computed by means of the Akaike Information Criterion.
Experimental data
File name: Test_x0.xx_y0.yyFs2MHz_1x1x0.003_Al.csv
Experimental data collected with custom AE instrumentation as described in Ref 1.
One single file is a collection of 3 tests (three repetitions of the impact event at the same position), each of them containing three signals acquired simultaneously by three sensors located in proximity of three corners of a 1x1x0.003 aluminum plate having the same geometrical and numerical characteristics of the numerical one. The sensors acquire 5000 samples at a rate of 2 MHz (indicated in the file name) with a pre-trigger window of 1500 samples. The specific coordinates of the sensors are:
s1 [x = 0.05, y = 0.95] m
s2 [x = 0.05, y = 0.05] m
s3 [x = 0.95, y = 0.05] m
There are 9 files associated with as
本卡描述如含来自官方页的原文片段,其版权归原作者,不在本站 CC-BY 4.0 许可范围(见关于与许可)。
- 落地页
- https://zenodo.org/records/10875042
- 许可证
- CC-BY-4.0 (置信:verified_official)
- 发布年份
- 2024
- 发布方
- Zenodo
分发点
| zenodo | https://zenodo.org/records/10875042 |
关联论文(3 篇)
仅表示这篇论文与本数据集在文献网络中相邻, 不代表该论文确实使用了本数据集。
- A Tiny Machine Learning Approach to the Edge Localization of Acoustic Sources via Convolutional Neural Networks 候选引用(未核验)
- Tiny Deep Learning Architectures Enabling Sensor-Near Acoustic Data Processing and Defect Localization 候选引用(未核验)
- Deep Learning Approaches for Robust Time of Arrival Estimation in Acoustic Emission Monitoring 候选引用(未核验)
溯源(6 条)
| 来源链接: https://zenodo.org/records/10875042 日期: 2026-07-21 |
| 来源链接: https://api.datacite.org/dois/10.5281/zenodo.10875042 日期: 2026-07-30 |
| 来源链接: https://api.datacite.org/dois/10.5281/zenodo.10875042 日期: 2026-07-30 |
| 日期: 2026-07-30 |
| 日期: 2026-07-31 |
| 日期: 2026-07-31 |