# ml-arc-fault-detection

**Repository Path**: MicrochipTech/ml-arc-fault-detection

## Basic Information

- **Project Name**: ml-arc-fault-detection
- **Description**: A baseline model for Arc Fault Detection using an IEEE dataset
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-06-23
- **Last Updated**: 2026-07-11

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

<p align="center">
  <img src="images/mchpedgeaibanner.png" alt="Microchip Edge AI" width="100%">
</p>

![Status](https://img.shields.io/badge/status-active-brightgreen)
![Platforms](https://img.shields.io/badge/platforms-MCU%20%7C%20MPU%20%7C%20FPGA%20%7C%20dsPIC-red)
![Domain](https://img.shields.io/badge/Edge%20AI-%20Applications-orange)
![ML](https://img.shields.io/badge/Machine%20Learning-Embedded-yellow)
![License](https://img.shields.io/badge/license-Microchip-blue)

# ML Arc Fault Detection (AFD) Baseline Model
A baseline machine learning model for Arc Fault Detection demonstrating high accuracy and small memory footprint suitable for deployment on embedded systems.

## Motivation

Arc fault detection is a critical safety challenge in electrical systems, where undetected faults can lead to fires and equipment damage. Traditional rule-based methods struggle to distinguish genuine arc faults from normal switching transients and varying load conditions.

This repository demonstrates how **machine learning dramatically outperforms traditional approaches** while remaining practical for deployment on resource-constrained embedded devices. Using the MPLAB ML Development Suite and the publicly available IEEE dataset, we achieve:

- **100% classification accuracy** on benchmark data
- **Sub-millisecond inference** latency enabling real-time detection
- **Small memory footprint** suitable for microcontrollers
- **Production-ready** embedded deployment with simple C API

## Overview

This repository provides a complete workflow for building arc fault detection models using:
- **MPLAB ML Development Suite** - Automated ML pipeline for embedded systems
- **IEEE Dataset** - Publicly available labeled arc fault data
- **Pre-processed Data** - Cleaned datasets ready for training
- **Documentation** - Comprehensive guide from data to deployment

## Repository Contents

```
├── ieee-dataset/
│   └── ieee-dataset-cleaned.zip           # Compressed cleaned datasets
│       ├── cleaned_sequential_parallel_basic.csv    # Parallel AFD without load
│       ├── cleaned_sequential_parallel_load.csv     # Parallel AFD with load  
│       ├── cleaned_sequential_series_basic.csv      # Series AFD without load
│       └── cleaned_sequential_series_load.csv       # Series AFD with load
├── notebooks/
│   ├── ieee-data-cleaning.ipynb           # Data validation notebook
│   └── export-to-mplab-ml.ipynb           # MPLAB ML export notebook
├── AFD_MPLAB_ML_Complete_Guide.md         # This complete guide (Markdown version)
├── AFD_MPLAB_ML_Complete_Guide.pdf        # PDF version of the complete guide
└── README.md                              # Quick start guide and repository overview
```
## Quick Start

### 1. Get the IEEE Dataset
The cleaned datasets are included in this repository. The original data is available at: [IEEE DataPort - Low Voltage DC Series Arc Fault](https://ieee-dataport.org/)

### 2. Install MPLAB ML Development Suite
The MPLAB ML Development Suite is bundled for free as a plugin to the MPLAB X IDE which can be downloaded [here](https://www.microchip.com/en-us/tools-resources/develop/mplab-x-ide).  

A Pro version is also available, visit [MPLAB Machine Learning Development Suite](https://www.microchip.com/en-us/tools-resources/develop/mplab-machine-learning-development-suite) to learn more. 

### 3. Follow the AFD Complete Guide
Download [AFD_MPLAB_ML_Complete_Guide.pdf](AFD_MPLAB_ML_Complete_Guide.pdf) for step-by-step instructions covering:
- Data preparation and queries
- Feature extraction
- Model training
- Performance evaluation
- Embedded deployment

## Dataset Information

**IEEE Dataset Labels:**
| Label | Description |
|-------|-------------|
| -1    | Normal conditions, no arc present |
| 0     | Transient, typically at beginning/end of arc |
| 1     | Arc observed |

**Data Characteristics:**
- Sampling rate: 16 kHz
- Window size: 160 samples (10 ms)
- Four dataset variants: series/parallel × with/without load

**Data Quality:** The original IEEE dataset contained some invalid entries (NaN values, incorrect labels). Cleaning notebooks are provided, and the repository includes pre-cleaned versions ensuring reproducible results.

## Key Features for Arc Detection

The baseline model uses four carefully selected features providing optimal class separation:

1. **Statistical: Absolute Mean** - Average signal magnitude
2. **Statistical: Minimum** - Negative excursions characteristic of arcs
3. **Frequency: MFE** - Median frequency energy capturing spectral characteristics
4. **Time: Percent Time Over Zero** - Duty cycle and polarity information

These features enable **100% sensitivity** (no missed arc faults) and **100% specificity** (no false alarms).

## Model Performance

**Baseline Model Results:**
- ✅ 100% accuracy across all test datasets
- ✅ 100% sensitivity (no false negatives)
- ✅ 100% specificity (no false positives)
- ✅ Excellent generalization across series/parallel and loaded/unloaded scenarios
- ✅ Consistent performance on validation, training, and cross-validation sets

## Deployment

The trained model exports as a "knowledge pack" containing:
- Optimized C code for target processor
- Feature extraction routines
- Inference engine
- Simple integration API

**Example Integration:**
```c
#include "kb.h"

// Initialize at startup
kb_model_init();

// Run inference on 160-sample window
ret = kb_run_model((SENSOR_DATA_T *)data, num_sensors, KB_MODEL_rank_0_INDEX);

if (ret >= 0) {
    sml_output_results(KB_MODEL_rank_0_INDEX, ret);
    kb_reset_model(KB_MODEL_rank_0_INDEX);
}
```

**Resource Requirements:**
- Flash: ~4-6 KB (model + features)
- RAM: ~1-2 KB (inference workspace)
- Latency: <2 ms on typical dsPIC and MCUs

## Documentation

For complete details including:
- Motivation and background
- Step-by-step model building
- Feature analysis and visualization
- Performance evaluation
- Deployment instructions

**→ Download the [Complete Guide PDF](AFD_MPLAB_ML_Complete_Guide.pdf)**

## Hardware Recommendations

For production arc fault detection systems, consider processors with DSP capabilities:
- **dsPIC33A series** - Integrated DSP with 40 MSPS ADCs
- Any MCU with hardware MAC for accelerated feature extraction

## Notebooks

### ieee-data-cleaning.ipynb
- Identifies and removes invalid data entries
- Documents data quality issues found in original dataset
- Generates cleaned CSV files

### export-to-mplab-ml.ipynb  
- Demonstrates formatting data for MPLAB ML import
- Shows auto-labeling process
- Provides dataset statistics

## Contributing

This is a baseline model demonstrating the workflow. For your specific application:
1. Collect data from your hardware platform
2. Test under all expected operating conditions
3. Fine-tune features and models as needed


## Additional Resources

- [MPLAB ML Development Suite Documentation](https://onlinedocs.microchip.com/oxy/GUID-80D4088D-19D0-41E9-BE8D-7AE3BE021BBF-en-US-3/GUID-5C3084BA-4706-40CF-AD6E-6AC561A13DF9.html)
- [IEEE Research Paper](https://ieee-dataport.org/) - "Why AI: A Comparative Study for Detection Methods in DC Series Arc Fault"
- [Technical Support](https://microchip.com/support)

## License

Copyright © Microchip Technology Inc. All rights reserved.

## Why This Matters

Arc fault detection saves lives and prevents property damage. Machine learning enables more reliable detection than traditional methods while remaining practical for embedded deployment. This guide provides a proven starting point for developing production-ready arc fault detection systems.

---

**Ready to get started?** Download the [Complete Guide](AFD_MPLAB_ML_Complete_Guide.pdf) and follow along with the included datasets and notebooks!