HarmoniSense-LV

Low-Voltage Distribution Network Physical Fingerprint and AI Expert Diagnostic Platform

Python Version
License: CC BY-NC-SA 4.0
Build Status
Engine
Platform
HarmoniSense-LV System Dashboard Mastery

HarmoniSense-LV is an AI-driven diagnostic system for low-voltage grids based on high-order physical harmonic fingerprints. It enables automatic topology reconstruction of the physical distribution network via AI algorithms, performs multi-dimensional fingerprint identification of connected loads, and quickly locates electricity theft, unregistered users, and unauthorized connections of new energy equipment.

📖 Auxiliary Documentation

To facilitate developers and users, the project provides detailed Markdown documentation:


🚀 Core Features

  • AI Topology Reconstruction: Based on the Pearson correlation coefficient algorithm, it automatically reconstructs the physical phase and link structure of the substation area without manual record entry.
  • Load Fingerprint Identification: Deep feature extraction technology accurately identifies specific electricity consumption patterns such as EV charging, distributed photovoltaics, cryptocurrency mining machines, and high-power heat pumps.
  • Unauthorized Node Detection: Through physical energy conservation residual analysis, it locks onto unauthorized connection (electricity theft/unregistered user) nodes in seconds.
  • Multi-language Interaction: Built-in dynamic switching between Chinese and English, providing an industrial-grade real-time interactive dashboard.

🛠️ Technical Architecture

The project adopts a modular design with a clear code structure:

  • dashboard_app.pyMain program entry point. Built on Dash (Plotly), responsible for global state management and callback logic.
  • app_logic.pyAI core algorithm center. Contains data cleaning, simulation engine, phase identification, and anomaly analysis logic.
  • app_viz.pyTopology graph rendering engine. Responsible for NetworkX spatial computation and Plotly dynamic topology visualization.
  • app_components.pyUI component library. Encapsulates sidebar, cards, expert report box, and accordion components.
  • app_translations.pyInternationalization dictionary. Supports full business terminology mapping between Chinese and English.

🚦 Quick Start

1. Environment Setup

Ensure your Python environment supports the following libraries:

pip install dash dash-bootstrap-components pandas networkx numpy scipy openpyxl

2. Launch the System

Run the main script to start the Flask/Dash service:

python dashboard_app.py

After launching, visit: http://127.0.0.1:8053

📚 Theoretical Foundation & Acknowledgements

The algorithm design philosophy and physical mapping logic of this platform are deeply inspired by the following academic achievements:

  • Paper TitleUtilising Smart-Meter Harmonic Data for Low-Voltage Network Topology Identification
  • Core Team: Ali Othman, Neville R. Watson, Andrew Lapthorn (University of Canterbury); Radnya Mukhedkar (EPECentre).
  • Published JournalEnergies 2025, 18(13), 3333.
  • Paper Linkhttps://doi.org/10.3390/en18133333

Acknowledgements: Special thanks to the research team at the University of Canterbury for their pioneering work in the field of harmonic analysis for low-voltage distribution networks.


⚖️ License

This project is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).

  • You are free to: Share (copy and redistribute) and Adapt (remix, transform, and build upon) the material.
  • Under the following terms:
    • Attribution: You must give appropriate credit and provide a link to the license.
    • Non-CommercialYou may not use the material for commercial purposes.
    • Share-Alike: If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.

© Designed & Developed by Clark | I have obtained the key to Babel, and I shall raise countless towers in Shinar

滚动至顶部