Elastic Water Column Model (EWCM) with Dynamic Elements
This work is part of my PhD research at the University of Adelaide, focused on developing an advanced transient model for water distribution systems (WDSs). The EWCM provides a fast, controllable, and physics-based framework that incorporates dynamic components such as PRVs and pumps, supporting both operational analysis and real-time control in modern water networks.
What problem does this solve?
Utilities need to simulate both gradual operations (PRV setpoint changes) and sudden events (pump trips, fast valve operations) without long run times. EWCM handles both in one framework and scales efficiently on large networks.
What’s the outcome for operators?
- Preview pressure impacts of PRV changes and pump ramps before implementing.
- Identify vulnerable assets under fast transients and compare mitigation options.
- Use the state-space form for controller design, alarms, and scenario testing.
Interactive concept: Accuracy vs Runtime
The performance of the model can be tuned through discretization, unlike the Rigid Water Column Model. Its runtime is significantly faster than the Method of Characteristics, which is implemented in commercial tools such as HAMMER commonly used in the water industry.
How it works
Network graph
Pipes, PRVs, pumps are edges; nodes are junctions/reservoirs.
ODE-based EWCM
Elastic water column per pipe; accuracy increases with per-pipe “reaches”.
Dynamic elements
PRVs and pumps use actuator dynamics + algebraic head relations.
Show technical details
- Captures wave travel/fast events (vs RWCM) while remaining ODE/state-space (easier for control) compared to MOC.
- Reaches per pipe tune fidelity for fast transients.
- PRVs/pumps added as two-port dynamic components.
- Sparse matrices reduce runtime/memory on big systems.