The Alexandra Institute // Cases // CHAIN – Smart Water Networks

CHAIN – Smart Water Networks

Case

CHAIN – Smart Water Networks

Artificial intelligence to contribute to sustainable water management

The purpose of the CHAIN project is to develop innovative solutions for smart water networks based on artificial intelligence and machine learning. The objective is to find ways to increase efficiency, safeguard water resources and ensure security of supply.

The use of artificial intelligence, machine learning and data analytics is still at an early stage in the Danish water sector. However, there is an untapped potential for optimising the entire water supply chain if the water sector adopts smart technologies in combination with existing systems. This will not only have a positive impact on operational efficiency but also on sustainability.

In the strategic innovation project CHAIN, we collaborate with a number of partners to explore ways to make artificial intelligence available to the water sector.

In the long term, CHAIN will help water utilities operate more efficiently and enable them to predict and respond to challenges – even under extreme conditions. Consumers will continue to enjoy security of supply, and companies will be able to develop new services and products for the water sector, which will ultimately increase their market share and create new jobs.

Furthermore, CHAIN is closely linked to several of the United Nations' Sustainable Development Goals.

The CHAIN project is funded by Innovation Fund Denmark and is run by DHI.

The objective of CHAIN is to exploit the benefits of automation in the water supply infrastructure – from source to water utility to consumer.

Our main contribution to the project is machine learning and artificial intelligence. We develop standardised components that will be integrated with technologies developed by our project partners and tested in the utility companies in Skanderborg and Aarhus. The components developed during the project are based on existing methodologies and standardised tools in order to enable configuration for other installations later on.

More specifically, we will deliver a Machine Learning Library for water networks, which includes:

  • Recognition of operations
  • Consumption prognosis
  • Prognosis of sensor measurements

A major technological challenge in CHAIN is the diversity of data as well as the dynamics and complexity of a water supply network.

We use well-known machine learning techniques to deal with this challenge, e.g.:

  • Real-time data
  • Time series
  • Machine learning techniques
    • Training on data sets (supervised learning)
    • Unsupervised anomaly detection
  • Parameter optimisation
  • Hydrodynamic mathematical modelling

The innovative aspect of CHAIN is the attempt to introduce artificial intelligence in the water sector.

At present, there is only limited experience of using data-driven technologies to optimise the water supply infrastructure.

CHAIN will address the technological barriers associated with the use of machine learning, artificial intelligence and data analytics together with existing systems and tools in the water sector.

In CHAIN we collaborate with the following technology partners and water utility companies:

The total budget is DKK 8,8 million funded by Innovation Fund Denmark.

We can help you use digital technologies as an instrument to achieve sustainable business goals.

Contact
Manager – Intelligent Sustainability
People, Technology and Business Lab
+45 30 23 66 22
See profile
.