Automatic generation of water distribution systems based on GIS data

Area of Science:

• Hydraulic engineering
• Water resource management
• Geographic Information Systems (GIS)

Background:

• Case study research is crucial for evaluating water distribution system (WDS) optimization strategies and software.
• Limited data availability for real-world WDS analysis hinders research due to high collection and setup costs.

Purpose of the Study:

• To present a novel algorithm for generating synthetic WDSs from GIS data.
• To address the challenge of limited data availability for WDS research.
• To create realistic WDS models for testing and benchmarking.

Main Methods:

• Developed an algorithm to generate WDSs using GIS data, including population density, housing density, and elevation.
• Generated synthetic WDSs and compared their network properties and hydraulic performance against actual systems.

Main Results:

• Generated WDSs exhibit comparable network properties and hydraulic performance to actual systems.
• Pressure head differences in generated WDSs were ±4 m or less for 75% of the supply area compared to actual systems.
• The methodology can generate WDSs of varying complexity, excluding components like valves and pumps.

Conclusions:

• The new algorithm effectively generates realistic WDS models from GIS data, overcoming data scarcity.
• This approach facilitates the testing of design and optimization algorithms on numerous networks.
• The method can aid in estimating construction costs for planned WDSs and comparing expansion strategies.