PINN | Hugo Academic CV Theme

Research Project Planning & AI Experiments Management

Sat, 14 Jun 2025 00:00:00 +0000

Effectively manage your AI-driven research projects for smart water networks — from ideation to simulation, paper writing, and presentations.

Ideation (Mind Mapping)

Use Hugo Blox’s mindmap extension to brainstorm ideas for a new hybrid AI model or grant proposal (e.g. Horizon Europe or Global Water Summit initiative):

- AquaTwin+ Project
- Hybrid AI
- GNN
- PINN
- Fuzzy Logic
- SCADA Integration
- Pollution Modelling
- Leak Detection
- DMA Zoning
- Partners
- TUHH
- Hamburg Wasser
- КПІ
- Київводоканал
- Siemens
gantt
dateFormat YYYY-MM-DD
title Hybrid AI for Smart Water Networks
section Setup
Define Objectives :done, 2025-05-01, 5d
Prepare Data :done, 2025-05-06, 5d
section Development
GNN Modeling :active, 2025-05-11, 10d
PINN Integration :2025-05-21, 7d
Pollution Scenarios :2025-05-28, 5d
section Dissemination
Prepare GitHub Repo :2025-06-02, 2d
Global Summit Slides :2025-06-04, 2d
- [x] Import Walkerton EPANET model
- [x] Add base demand & chlorine decay config
- [x] Integrate PINN for unknown pipe zones
- [x] Train GNN on simulated leaks
- [ ] Implement RL agent for tariff control
- [ ] Deploy interactive dashboard (Streamlit/Leaflet)
- [ ] Prepare Horizon Europe proposal PDF

Preprint: Fuzzy-GNN and PINN-based Modeling of Water Infrastructure under SCADA Blackout Conditions

Tue, 10 Jun 2025 00:00:00 +0000

This preprint builds upon the results of our previous published work on GIS-ANFIS-based accident prediction in water networks.

Due to the critical nature of infrastructure data and the wartime conditions in Ukraine, full access to source code and datasets is restricted. Only general architecture and anonymized examples are shared publicly.

The proposed model includes:

Fuzzy membership layers on top of graph node attributes
Integration of PINN for physical constraint modeling (e.g., Darcy–Weisbach)
Compensation mechanisms for missing SCADA sensor inputs
NSGA-II multi-objective optimization for valve control

This approach is applicable in utility AI systems that must remain resilient even during cyber-physical or war-induced infrastructure disruptions.

Adaptive AI for Smart Water Networks

Wed, 14 May 2025 00:00:00 +0000

Welcome to my research blog

Table of Contents

Vision

Water is life. Yet in many regions, aging infrastructure, climate change, and unpredictable consumer behavior challenge our ability to deliver it sustainably.

My mission is to develop adaptive AI systems that empower utilities, researchers, and communities to:

detect leaks early,
optimize hydraulic networks,
model pollution and chlorination dynamics,
and adapt in real time to behavioral disturbances.

Research Topics

Hybrid Graph Neural Networks (GNN) + Physics-Informed Neural Networks (PINN) for real-time simulation and anomaly detection in water grids
Fractal geometry & metaheuristics to optimize valve placement and zoning (DMA)
SCADA blackout simulation using I-PINN + synthetic data
Water quality modeling including disinfectant decay and pollutant tracking
Multi-agent systems with RL for tariff-policy simulation and response
GeoAI and vector databases for smart urban water analytics (Neo4j + PostGIS + SCADA + LangChain)

Get Involved

Tools & Methods

Python stack: PyTorch Geometric, DeepXDE, pymoo, Streamlit, WNTR, epanet-python
Databases: Neo4j, PostGIS, TimescaleDB, Vector DBs for AI
Frameworks: Hugo Blox + GitHub Pages
Deployment: Docker, API-driven services for interactive dashboards

Themes & Publishing

This site is powered by Hugo Blox, with fully customizable design and automated publishing. All content is written in Markdown + YAML for transparency and simplicity.

Credits

This site template is based on Hugo Blox
Released under the MIT License.

Hybrid AI Models for Adaptive Water Infrastructure: GNN and PINN Integration with SCADA Signals

Tue, 10 Dec 2024 00:00:00 +0000

Click the Cite button above to export the BibTeX and add this to your reference manager or Zotero.

These slides provide an overview of the SCADA-integrated PINN-GNN architecture for water supply optimization.

The proposed approach demonstrates how hybrid neural architectures can be applied to real-world water infrastructure challenges. By combining graph topologies with physical modeling constraints and real-time sensor feedback, the framework bridges data-driven and physics-based modeling.

It supports operational decisions such as:

Dynamic leak localization
Adaptive pressure control
Placement of smart valves and sensors
Predictive failure analysis

Future work includes deployment of the system in collaboration with water utilities across Europe within the Horizon Europe framework.