Drops of knowledge

Customer experience, Digital technologies, Urban water, Water utilities

How real-time modelling supports a proactive water network operation

With rapid urban population growth, water utilities worldwide are more than ever facing pressing challenges to preserve the hydraulic perfomance and water quality of already strained water distribution networks and infrastructure. Basically, water distribution system’s primary role is to ensure a safe and reliable water supply to consumers, while maintaining efficiency at a low cost. This strain of growing populations and aging infrastructure has created an increasing need for implementing real-time water network management systems to support proactive management of water distribution networks. Such smart management systems are typically using advanced technology like Digital Twin modelling.

A Digital Twin is a digital replica of a living physical entity. In various industrial sectors, Digital Twins are being used to optimise the operation and maintenance of physical assets, systems and processes. In the context of water distribution systems, a hydraulic model that is linked to the telemetry system provides a Digital Twin of the physical water distribution network and allows for detailed analysis of the system conditions in real-time.

Exploiting real-time modelling for better efficiency, economy and consumer service

By implementing real-time control using Digital Twin modelling, operators are provided with operational decision support. Real-time monitoring systems periodically and automatically read real-time or historical data from an exchange database, such as the Supervisory Control and Data Acquisition (SCADA) database, update the hydraulic model, run the hydraulic simulation, and provide results including hydraulic decisions, alarms and parameter settings. Which means that network failures can be detected at an early stage and let operators react quickly to minimise damaging effects. In this way Digital Twin modelling supports a proactive water network operation, resulting in more economically efficient networks, faster response to system failures, and improved consumer service.

Read on to discover how efficient network modelling, real-time simulations, as well as hindcasting and forecasting capabilities, all contribute to proactive real-time monitoring and control of water distribution systems.

Network modelling and simulation

By using modelling and simulations it is possible to provide a complete online water distribution hydraulic modelling system and perform analysis of the water network. Simulations can be automatically and periodically executed with real-time data from a SCADA monitoring system. The hydraulic simulations include steady state, extended period, and hydraulic transient analyses of transmission lines and/or distribution networks. In this way the simulations can predict possible problems by using predefined or user-defined alarms and alert operators if events are predicted to occur. In addition, complications such as outages of pumps, valves, reservoir shutdowns or pipe failures can be rapidly accessed and allow system operators to find remedial actions.

Respond quickly to changing conditions

In a real-time modelling system, data is received from the SCADA monitoring database and is used as input to the model, the simulations are then run and perform an immediate analysis of the system status and allow operators to assess potential problems respond quickly to changing conditions. In this mode, the system operates in a continuous cycle of predefined time steps. During each cycle, a hydraulic and water quality analysis is performed, and the model parameters are updated with measured SCADA data. Subsequently, the output data from the model can be displayed in a dashboard alongside data from the SCADA process monitor.

Operation based on hindcast modelling

Hindcast modelling can be performed for any past event within a defined period of time – for example, the last 6 hours or the last week. Hindcast modelling allows the system to evaluate and determine node demands and diurnal curves. Moreover, control rules can be established to change pipe status or settings based on tank water levels, junction pressure, or time of day. Even complex rules based on a list of parameters available in the SCADA database can be reproduced. Hindcast modelling can also be used to populate complex models with real SCADA data, avoiding time-consuming data handling for model configuration.

Manage settings by forecasting demand

The behaviour of a water distribution system can be short-forecasted within selected time intervals – for example, 6 or 12 hours. Conditional control rules serve to adjust link settings or status, and these rules can either be predefined and adjusted based on time, or developed in hindcast mode. Using forecast modelling, the model offers various options for handling node demands, such as reading them from the SCADA system or adjusting a predefined node demand based on the measured inflow – depending on the cause of the demand.

Curious to learn more?

Of all the different types of wet infrastructure and systems, water distribution systems are the most fundamental to sustaining our quality of life in modern society. Therefore, securing optimal operation of water distribution systems is of critical importance. Learn more about digital simulations and discover how you can proactively manage your daily water network operations with applications such as the WaterNet Advisor.

About the authors

Petr Ingeduld, Senior Project Manager and Hydraulic Engineer

Petr is a water supply systems expert at DHI with over 20 years of professional experience. Throughout his career, Petr has focused on application development of hydraulic models to support water supply and distribution. He has managed numerous global projects and provided technical direction and troubleshooting on various modelling efforts.

Davide Persi, Senior Project Manager and Hydraulic Engineer

Davide is a fluvial and urban networks expert at DHI with 20 years of professional experience. Davide has been involved in several projects as a modeller or project manager with focus on flooding management, morphological analysis, water supply and distribution optimisation as well as drainage design and rehabilitation. Recently he has managed several projects focused on the implementation of DSS systems.