For decades, the US Environmental Protection Agency (EPA) has provided EPANET as the global industry standard for modeling pressurized water distribution networks. However, as modern smart-water networks demand advanced data science workflows, machine learning integrations, and rapid prototyping, traditional desktop interfaces fall short. Developed by the international research initiative WaterFutures , bridges this gap, allowing civil engineers, researchers, and data scientists to programmatically build, simulate, and optimize complex water infrastructure directly within code-driven ecosystems. Key Capabilities of the Core Engine
It works seamlessly with EPANET-MSX, allowing for the simulation of complex chemical reactions beyond just chlorine residuals (e.g., disinfectant byproducts, blending, and advanced chemistry).
Packages entire simulated scenarios into shareable assets or streams them directly to remote tracking frameworks via web interfaces. 2. EPyT-Control and Cyber-Physical Systems epanet plus
EPANET Plus addresses these issues by rewriting and extending the engine while maintaining full compatibility with the standard .inp file format.
: Serves as the foundation for the EPyT-Flow package, providing a high-level Python interface for researchers to generate hydraulic and water quality scenario data. For decades, the US Environmental Protection Agency (EPA)
: Earlier versions of the software were notable for fixing persistent bugs in the original EPANET 2.0, such as the incorrect orientation of flow arrows and errors during the export of specific file types.
Supports complex rule-based controls for pump and tank management. Key Capabilities of the Core Engine It works
d.runSimulation()
At its heart is the EPANET engine , an open-source industry standard for modeling water distribution systems. While the base EPA version is free and highly reliable, it lacks modern GIS integration and real-time capabilities. "Plus" versions—like , EPANET-JS , and commercial wrappers—bridge this gap. Key Features & Capabilities
: Using PDD, a utility can simulate a 24-hour electrical blackout. Tanks drain, pressures drop, and EPANET Plus shows exactly which neighborhoods lose water and at what time—not just reduced flow.
Building directly upon EPANET-PLUS, the EPyT-Flow package is designed for machine learning researchers and data scientists. It provides high-level objects to rapidly prototype massive data scenarios.