INTEREST: rules of thumb for costs and returns of sensor networks in drinking water distribution

Sensor networks allow us to monitor drinking water more comprehensively and detect contaminants more effectively. However, we need more practical experience to increase our knowledge in the area. Within the TKI Water Technology INTEREST project, the water companies Dunea, Evides, Vitens and WMD have worked with sensor developer, Optiqua Technologies, and the knowledge institute, KWR, on using simulations to develop rules of thumb for the relation between investments in and returns from sensor networks in different types of distribution networks. These rules of thumb allow for better-founded choices about the application of sensor networks, and possibly also lower the threshold for actual investments in sensor networks.

Better monitoring with sensors

The current monitoring of our drinking water quality consists primarily of random sampling of the pure water as it leaves the production location and as it flows from the customer’s tap, followed by lab analysis of limited water volumes. Now that more and more water-quality sensors are available on the market, it is possible to significantly improve monitoring coverage, both in terms of time and space. But experimentation with the application of sensors in networks is still very limited, because there is little clarity about the benefit they might provide. This holds back innovation in the field of sensors and the improvement of water-quality monitoring.

Rules of thumb for sensor networks

Recent research within the framework of the water sector’s joint research programme has produced a method which permits the quantification of a sensor network’s performance, and connects it to the investment required to achieve said performance. Using numerical simulations, it is possible to get a picture of the performance of sensor networks without having to actually install them. The project partners conducted such simulations for a number of representative network types (urban, rural, mainly-meshed, mainly-branched, and combinations), using varying numbers of water-quality sensors, placed in optimal locations to detect contaminants. In this way, the relationship between the number of sensors in a network and the contamination detection probability in that network can be quantified. By also incorporating the results of a cost model for Eventlab sensors for general water quality, a relation can be established between investment and performance. The results provided the basis for the rules of thumb for different types of networks and their performance. For example, for urban distribution networks, the detection probability is higher because these networks, per kilometre of pipe, supply large volumes of water and have a higher degree of meshing. The installation of sensor networks involves not only the costs of the sensor hardware, but also those related to setting up monitoring stations, and sensor maintenance and replacement. Higher performance levels require larger numbers of sensors, while as the number of sensors increases the performance increases more slowly than do the costs. By increasing scale[JA1] , using monitoring stations for several sensors, and innovating sensor technology one could, however, produce a more favourable cost-benefit ratio in the future.

Building knowledge and experience

The developed rules of thumb make it easier for water companies and sensor suppliers to make well-founded choices regarding (pilot) projects with sensor networks, and this will contribute to further innovation. It is important that these rules of thumb be regularly updated with new information from projects of this type. Water companies, sensor suppliers and KWR can learn from each other by sharing their knowledge and experience when it comes to investments in and the performance of sensor networks; the rules of thumb are a useful tool in this regard. Read more about this project on the TKI Water Technology website.