WaterModels.jl Documentation


WaterModels.jl is a Julia/JuMP package for steady-state potable water distribution network optimization. It is designed to enable the computational evaluation of historical and emerging water network optimization formulations and algorithms using a common platform. The code is specifically engineered to decouple Problem Specifications (e.g., water flow, optimal water flow, network design) from Network Formulations (e.g., mixed-integer linear, mixed-integer nonlinear). This decoupling enables the definition of a wide variety of water network optimization formulations and their comparison across several common problem specifications.


The latest stable release of WaterModels can be installed using the Julia package manager with

] add WaterModels

For the current development version, install the package using

] add WaterModels#master

Finally, test that the package works as expected by executing

] test WaterModels

Usage at a Glance

At least one optimization solver is required to run WaterModels. The solver selected typically depends on the type of problem formulation being employed. As an example, to solve a mixed-integer linear programming (MILP) formulation of the feasible water flow (wf) problem, the open-source MILP solver HiGHS can be used. Installation of the JuMP interface to HiGHS can be performed via the Julia package manager, i.e.,

] add HiGHS

Then, as one example, a piecewise-linear, relaxation-based convexification of the physics for the well-known Shamir (two-loop) network, using an error tolerance of one meter to model the envelope of each pipe's Hazen-Williams head loss curve, can be solved to feasibility using

using WaterModels, HiGHS

# Parse the network data from an EPANET file.
network = parse_file("examples/data/epanet/shamir.inp")

# Set linearization partitioning points that assume a head loss error tolerance of one
# meter, with widths between flow points no greater than 1.0e-4 cubic meters per second.
set_flow_partitions_si!(network, 1.0, 1.0e-4)

# Solve the corresponding relaxation of the water flow problem.
result = solve_wf(network, PWLRDWaterModel, HiGHS.Optimizer)

After solving the problem, results can then be analyzed, e.g.,

# The termination status of the optimization solver.

# Transform solution data to SI units.

# The flow along pipe 4 in cubic meters per second.

# The total hydraulic head at node 2 in meters.

# The pressure head at node 2 in meters.