Smart water solutions

The water sector faces significant challenges around water quantity and quality. Water utilities and Water Authorities must address emerging contaminants, the effects of climate change, stricter discharge limits and regulations, and sustainability goals, such as climate neutrality or carbon footprint reduction.

Methods, tools and products

At KWR, we develop smart water solutions that improve the design, management and performance of water systems.  We offer expertise in process and mechanistic modelling, artificial intelligence (AI), optimisation, uncertainty studies, real-time control and data visualisation. By combining our strong expertise in the water domain with enabling digital technologies, we provide innovative and smart water solutions that help solve challenges and contribute to optimal water management, today and in the future.

Smart monitoring

KWR conducts advanced smart monitoring campaigns, using IoT sensors and online analysers, enhanced with a robust cloud-based data acquisition system, to measure variables that are difficult to acquire, such as nitrous oxide (N2O) emissions from wastewater treatment plants. These unique data are essential for the quantification of N2O emissions – a potent greenhouse gas – and support modelling efforts to conceptualise mitigation measures. Additionally, we develop tailored, advanced visualisation and decision-support systems for end-users,  powered by the use of online sensor data and conducting data analytics. For instance, we have created a user-friendly decision-support system to increase the efficiency of water recirculation in greenhouse horticulture.

Applications throughout the water cycle 

We provide smart water solutions that cover the entire water cycle:

• Process-based and mechanistic modelling for drinking water treatment steps, such as coagulation, activated carbon and advanced oxidation.
• Design of pipe and sensor networks for robust and future-proof distribution of drinking water.
• Biokinetic process modelling for activated sludge treatment in wastewater treatment plants – including a particular focus on emerging challenges such as organic micropollutant (OMP) removal and N2O emission reduction.

Data-driven approach and AI

In addition, KWR is developing data-driven approaches and applications of artificial intelligence for a spectrum of use cases related to water and wastewater. We use machine learning modelling for the identification and counting of microplastics in samples. We also apply machine learning to better understand N2O production and predict its emissions. AI applications, coupled with process knowledge, are being developed to support and implement control and mitigation measures that help reduce N2O emissions. KWR is also researching the use of machine learning models for the purpose of water demand forecasting, and developing reinforcement learning methodologies aimed at supporting the more flexible design of urban water distribution systems.

AquaPriori

New substances that need to be identified and removed turn up regularly in the water cycle. For this, quick assessment methods are needed to help determine whether, and how, these substances can be removed. Our tool AquaPriori is tailored for this task, and provides a multifaceted web application for different drinking-water and wastewater treatment technologies.

Numerical optimisation of drinking water distribution systems 

With the software platform Gondwana, drinking water utilities can optimise their drinking water distribution systems. The platform combines a very flexible and multifaceted approach to the definition of optimisation issues with a user-friendly graphic interface.

• Examples of application possibilities Gondwana include:
  design of target structures and transitions from current configurations to target structures, with low cost and good performance as objectives. Performance can be determined by, among other things, the degree of security of supply, desired pressure, water quality, energy consumption and number of failures;
• determining optimal DMAs (District Metered Areas), to maximise the detectability of leaks and other anomalies with as few volume flow meters as possible;
• determining optimal valve configurations, with the objectives of low sub-standard delivery minutes (OLM) and good spillability of sections;
• determining optimal locations of water quality sensors or samplers, with the objectives of minimising detection time, maximising capture probability or coverage, facilitating source determination and more;

In all applications, different scenarios can be taken into account, such as changes in water demand, outages and contamination.