AFCI offline kit - Part 11: MIT arc-fault current recordings 1/2 (burner, fan)
Offline development kit for an arc-fault detection / electrical-fire prevention system (current sensor -> STM32F407 ADC+DMA -> FFT/STFT features -> quantized DNN -> trip). Target offline PC: Windows 10/11 x64, CPU-only Python training.
Contents: MIT arc-fault current recordings (1/2): burner & fan
Folder MIT_Siegel2018_Burner_Fan/ (6 WAV files, 1.55 GB).
About this dataset (Parts 11, 12)
MIT (Siegel et al. 2018) — "Electronic Circuit Current Data", training set for Real-time Deep Neural Networks for internet-enabled arc-fault detection (J. E. Siegel, S. Pratt, Y. Sun, S. E. Sarma, Engineering Applications of AI, 2018). Current-sensor recordings stored as 48 kHz, 16-bit mono WAV, four device classes (burner, fan, iMac, ozone generator) under normal and arcing conditions. Read with scipy.io.wavfile.read or soundfile.
Source: Harvard Dataverse, doi:10.7910/DVN/IFDIZ1 — license CC-BY-NC-4.0 (non-commercial use only; cite the authors).
| Part | Files |
|---|---|
| 11 | Burner.wav, Burner2.wav, Fan.wav, Fan2.wav, Fan3.wav, Fan4.wav |
| 12 | iMac1.wav, iMac2.wav, Ozone.wav, Ozone2.wav, Ozone3.wav |
Which recordings are "normal" vs "arcing" is described in the paper/Dataverse page; the thesis PDF could not be fetched automatically (MIT DSpace), so get it from https://dspace.mit.edu (search "Real-time Deep Neural Networks for internet-enabled arc-fault detection") or the journal.
Note: 48 kHz is close to what an STM32F407 can sample comfortably, so this set is useful for checking how well a low-sample-rate model works.
Rebuilding split files
Files over 100 MB were split into *.part001, *.part002, ... pieces. After downloading the whole repo:
- Windows:
powershell -ExecutionPolicy Bypass -File .\join_parts.ps1 - Linux/macOS:
bash join_parts.sh
Both scripts rebuild the files and check them against SHA256SUMS.txt.
Files (7 files, 1.4 GB total)
| File | Size | Split into |
|---|---|---|
download_log_stage11.json |
1.0 KB | - |
MIT_Siegel2018_Burner_Fan/Burner.wav |
82.4 MB | - |
MIT_Siegel2018_Burner_Fan/Burner2.wav |
329.6 MB | 4 |
MIT_Siegel2018_Burner_Fan/Fan.wav |
82.4 MB | - |
MIT_Siegel2018_Burner_Fan/Fan2.wav |
329.6 MB | 4 |
MIT_Siegel2018_Burner_Fan/Fan3.wav |
329.6 MB | 4 |
MIT_Siegel2018_Burner_Fan/Fan4.wav |
329.6 MB | 4 |
Each file keeps the license of its original publisher (see the notes above).
Master index
This is one of 19 repos (AFCI_part1 … AFCI_part19). Overview of all parts: AFCI_part6 / 00_START_HERE_MASTER_INDEX.md · Installation: AFCI_part6 / 00_INSTALLATION_GUIDE.md