Integrated approach for optimal sensor placement and state estimation: A case study on water distribution networks

Area of Science:

• Engineering
• Environmental Science
• Data Science

Background:

• Water distribution networks (WDNs) require real-time monitoring of state variables like pressure and flow for optimal operation.
• Complete monitoring is often impractical due to cost and accessibility, necessitating efficient sensor placement and state estimation.

Purpose of the Study:

• To develop a hybrid approach for sensor placement (SP) and state estimation (SE) in WDNs.
• To enable accurate situational awareness of system state variables throughout the network.
• To improve the adaptive operation and management of WDNs.

Main Methods:

• A hybrid approach combining data correlation structures and fluid dynamics principles for SP and SE.
• Formulating the SE problem as a constrained matrix completion task.
• Utilizing benchmark WDNs for validation.

Main Results:

• Achieved a mean absolute percentage error (MAPE) below 5% for estimating nodal head.
• Successfully mapped the WDN state using a completed matrix.
• Demonstrated the feasibility of the hybrid SP and SE approach on benchmark networks.

Conclusions:

• The proposed hybrid method provides a practical solution for achieving situational awareness in WDNs.
• Accurate state estimation allows for more adaptive and efficient network operation.
• This approach can significantly enhance water management strategies.