Artificial intelligence ensures the quality of drinking water

In times of longer dry periods, the timely detection of leaking drinking water pipes becomes increasingly important. A project has now been developed at the Software Competence Center Hagenberg (SCCH) in which intelligent camera systems can minimize water loss in a water distribution network. The title: “Automated remote inspection of critical infrastructure using intelligent camera systems” (ARIKI).

Smart water networks

“In Austria, eight percent of water losses are due to leaks in outdated pipes. The electrical energy for pumping water contributes significantly to the total operating costs of the water supply system,” says Alexander Valentinitsch, who leads the ARIKI project. Algorithms check the water flow, the water pressure and the pump performance. Virtual sensors are used to locate leaks, which was previously very difficult to do due to a lack of sensors. “Unfortunately, there is currently no practical, secure and interpretable AI solution for intelligent water networks that also takes data protection into account. With our experience in the area of ​​ethical AI, we are striving to establish an intelligent and confidential water data system in Austria,” says Valentinitsch . He refers to an AI that was developed at the Hagenberg Software Competence Center for an intelligent water supply system for major Chinese cities. In the current project, project partners from Germany also specialized in intelligent sensor technology – primarily camera systems – in order to further develop AI in supply networks together with the team from Hagenberg. Valentinitsch: “With this bilateral cooperation, we hope that sustainable and energy-efficient water supply in Europe will be strengthened.”

The data remains with the water supplier. A system is being developed for utilities that enables ethically responsible and energy-efficient action in intelligent water networks. The water suppliers themselves take over the training of the data that is crucial for learning the algorithm. Only the learning results of the algorithm are transferred and merged into a common central model. Smaller communities in particular would benefit from this federal approach in terms of water supply. The project will develop innovative methods that enable secure and confidential water data systems and supply systems and that comply with legal and ethical data protection principles.

Looking for project partners

Project manager Alexander Valentinitsch is still looking for water suppliers, municipalities and pump manufacturers who would like to take part in this innovative project:
alexander.valentinitsch@scch.at

info By clicking on the icon you can add the keyword to your topics.

info
By clicking on the icon you open your “my topics” page. They have of 15 keywords saved and would have to remove keywords.

info By clicking on the icon you can remove the keyword from your topics.

Add the topic to your topics.