This project is a work in progress. Currently in the Phase 1: Mechanistic Engine (ALBATwin).
ALBATwin-Hub is a high-fidelity Digital Twin for microalgae-bacteria consortia in High Rate Algal Ponds (HRAP). It implements the ALBA model (Casagli et al., 2021) using a hybrid approach that combines rigorous mechanistic equations with Physics-Informed Neural Networks (PINNs).
- Mechanistic Engine: 19 coupled ODEs/DAEs simulating biological and chemical dynamics (pH, Gas Transfer, Light).
- Hybrid Modeling: Integration of physical laws into Deep Learning via PINNs for robust predictions under noise.
- Modern Stack: Managed with
uvfor lightning-fast dependency handling and reproducibility. - Scenario Analysis: Interactive simulation of "What-if" scenarios (weather impact, aeration control).
For a more detailed description of the system, please refer to the ARCHITECTURE and MATH_MODEL pages.
ALBATwin-Hub/
├── docs/
│ ├── development/
│ │ └── DEVLOG.md # 📝 Chronological log of technical decisions and daily progress
│ ├── ARCHITECTURE.md # 🏗️ System design, component interaction, and data flow diagrams
│ ├── MATH_MODEL.md # ➗ Detailed ODE/DAE equations and biological kinetics (ALBA model)
│ ├── PRD.md # 📋 Core objectives, user stories, and functional requirements
│ ├── REFERENCES.md # 📚 Scientific bibliography and data sources
│ └── ROADMAP.md # 🗺️ High-level project phases and long-term goals
├── .gitignore
├── pyproject.toml # ⚙️ Project metadata and dependency management
├── uv.lock # 🔒 Locked dependencies
└── README.md # 📖 Project overview
To see the development phases, please refer to the ROADMAP.
The main reference paper for the MVP of the project is:
- Casagli et al. (2021). ALBA: A comprehensive growth model to optimize algae-bacteria wastewater treatment in raceway ponds. Water Research, 190, 116734. https://doi.org/10.1016/j.watres.2020.116734
For the complete list of references, please refer to the REFERENCES page.
Made with ❤️ by Anibal Rojo