No abstract

Elsevier Brentan, BM.; Luvizotto, E.; Herrera Fernández, AM.; Izquierdo Sebastián, J.; Pérez García, R. (2017). Hybrid regression model for near real-time urban water demand forecasting. AbstractThe most important factor in planning and operating water distribution systems is satisfying consumer demand. This means continuously providing users with quality water in adequate volumes at reasonable pressure, thus ensuring reliable water distribution. During the last years, the application of Statistical, Machine Learning and Artificial Intelligence methodologies has been fostered for water demand forecasting. However, there is still room for improvement and new challenges concerning to on-line predictive models for water demand have aroused. This work proposes applying support vector regression, as one of the currently better Machine Learning options for short-term water demand forecasting, to build a base prediction. Over this model, a Fourier time series process is built to improve the base prediction. This addition produces a tool able to get rid of part of the errors and bias inherent to a fixed regression structure in response to new incoming time series data. The final hybrid process is validated using demand data from a water utility in Franca, Brazil. Our model, being a near real-time model for water demand, may be directly exploited in water management decision-making processes.

Management of large water distribution systems can be improved by dividing their networks into so-called district metered areas (DMAs). However, such divisions must be based on appropriated technical criteria. Considering the importance of deeply understanding the relationship between DMA creation and these criteria, this work proposes a performance analysis of DMA generation that takes into account such indicators as resilience index, demand similarity, pressure uniformity, water age (and thus water quality), solution implantation costs, and electrical consumption. To cope with the complexity of the problem, suitable mathematical techniques are proposed in this paper. We use a social community detection technique to define the sectors, and then a multilevel particle swarm optimization approach is applied to find the optimal placement and operating point of the necessary devices. The results obtained by implementing the methodology in a real water supply network show its validity and the meaningful influence on the final result of, especially, elevation and pipe length.

With the advance of new technologies and emergence of the concept of the smart city, there has been a dramatic increase in available information. Water distribution systems (WDSs) in which databases can be updated every few minutes are no exception. Suitable techniques to evaluate available information and produce optimized responses are necessary for planning, operation, and management. This can help identify critical characteristics, such as leakage patterns, pipes to be replaced, and other features. This paper presents a clustering method based on self-organizing maps coupled with k-means algorithms to achieve groups that can be easily labeled and used for WDS decision-making. Three case-studies are presented, namely a classification of Brazilian cities in terms of their water utilities; district metered area creation to improve pressure control; and transient pressure signal analysis to identify burst pipes. In the three cases, this hybrid technique produces excellent results.

Water supply utilities need to properly manage their systems to guarantee a quality supply. One way to manage large systems is through division into district metered areas (DMAs). Graph clustering with an unknown number of subdivisions, as in social network theory, has proven highly efficient in this sectorization problem. Several physical and hydraulic features may easily be used as criteria to suitably divide the network. In this work, we use social network community detection algorithms to define several DMA scenarios. Configurations mainly depend on nodal demand and elevation, but adaptations may be needed to guarantee full supply in future scenarios related to system growth-and rehabilitation actions may also be required. The problem associated with pipes and valves is first solved with three optimization methods. The best solutions then enter a new optimization process, where tank dimensions and valve set points are defined. This complex optimization-segregation approach enables an improvement in the hydraulic efficiency of the E-town network at an affordable cost, and this approach also determines the measures needed to meet the dry season requirements.

No abstract

No abstract

The new environmental paradigms imposed by climate change and urbanization processes are leading cities to rethink urban management services. Propelled by technological development and the internet of things, an increasingly smart management of cities has favored the emergence of a new research field, namely the smart city. Included in this new way of considering cities, smart water systems are emerging for the planning, operating, and managing of water distribution networks (WDNs) with maximum efficiency derived from the application of data analysis and other information technology tools. Considering the possibility of improving WDN operation using available demand data, this work proposes a hybrid and near real-time optimization algorithm to jointly manage pumps and pressure reducing valves for maximum operational efficiency. A near real-time demand forecasting model is coupled with an optimization algorithm that updates in real time the water demand of the hydraulic model and can be used to define optimal operations. The D-town WDN is used to validate the proposal. The number of control devices in this WDN makes real-time control especially complex. To cope with this feature, computational methods must be carefully selected and tuned. In addition to energy savings of around 50%, the methodology proposed in this paper enables an efficient system pressure management, leading to significant leakage reduction.