Abstract. Nowadays, drinking water utilities need an acute comprehension of the water demand on their distribution network, in order to efficiently operate the optimization of resources, manage billing and propose new customer services. With the emergence of smart grids, based on automated meter reading (AMR), a better understanding of the consumption modes is now accessible for smart cities with more granularities. In this context, this paper evaluates a novel methodology for identifying relevant usage profiles from the water consumption data produced by smart meters. The methodology is fully data-driven using the consumption time series which are seen as functions or curves observed with an hourly time step. First, a Fourier-based additive time series decomposition model is introduced to extract seasonal patterns from time series. These patterns are intended to represent the customer habits in terms of water consumption. Two functional clustering approaches are then used to classify the extracted seasonal patterns: the functional version of K-means, and the Fourier REgression Mixture (FReMix) model. The K-means approach produces a hard segmentation and K representative prototypes. On the other hand, the FReMix is a generative model and also produces K profiles as well as a soft segmentation based on the posterior probabilities. The proposed approach is applied to a smart grid deployed on the largest water distribution network (WDN) in France. The two clustering strategies are evaluated and compared. Finally, a realistic interpretation of the consumption habits is given for each cluster. The extensive experiments and the qualitative interpretation of the resulting clusters allow one to highlight the effectiveness of the proposed methodology.

No abstract

In France, the system of performance indicators defined within the relevant regulations does not include indicators relating to pressure, or the number of connections. So, Irstea and Veolia Water decided to create a new system of water loss indicators designed for use in France and that would take into account these parameters. The study was initially carried out at District Metered Area (DMA) level. These analyses showed the advantages of considering the ratio of losses to number of connections, but did not highlight a significant link between water loss levels and average pressure. After that, using a module that automatically calculates average pressure based on an EPAnet model of a network, it was possible to carry out a study at the system level. The results at the system scale were similar to those obtained at the DMA level. The study demonstrates that it is relevant to use a performance indicator of losses per connection. On the other hand, it showed no link between the volume of losses and the average pressure, which is contrary to what is commonly admitted.

Nowadays, drinking water utilities need an acute comprehension of the water demand on their distribution network, in order to efficiently operate the optimization of resources, the management of billing and to propose new customer services. With the emergence of smart grids, based on Automated Meter Reading (AMR), a better understanding of the consumption modes is now accessible for smart cities with more granularities. In this context, this paper evaluates a novel methodology for identifying 5 relevant usage profiles from the water consumption data produced by smart meters. The methodology is fully data-driven using the consumption time series which are seen as functions or curves observed with an hourly time step. First, a Fourierbased additive time series decomposition model is introduced to extract seasonal patterns from time series. These patterns are intended to represent the customer habits in terms of water consumption. Two functional clustering approaches are then used to classify the extracted seasonal patterns: the functional version of K-means, and the Fourier REgression Mixture (FReMix) 10 model. The K-means approach produces a hard segmentation and K representative prototypes. On the other hand, the FReMix is a generative model and produces also K profiles as well as a soft segmentation based on the posterior probabilities. The proposed approach is applied to a smart grid deployed on the largest Water Distribution Network (WDN) in France. The two clustering strategies are evaluated and compared. Finally, a realistic interpretation of the consumption habits is given for each cluster. The extensive experiments and the qualitative interpretation of the resulting clusters allow to highlight the effectiveness 15 of the proposed methodology.

No abstract