ALBERT

Description: Rainfall-runoff models can be classified into three types: physically based models, conceptual models, and empirical models. In this latter class of models, the catchment is considered as a black box, without any reference to the internal processes that control the transformation of rainfall to runoff. In recent years, some models derived from studies on artificial intelligence have found increasing use. Among these, particular attention should be paid to Support Vector Machines (SVMs). This paper shows a comparative study of rainfall-runoff modeling between a SVM-based approach and the EPA’s Storm Water Management Model (SWMM). The SVM is applied in the variant called Support Vector regression (SVR). Two different experimental basins located in the north of Italy have been considered as case studies. Two criteria have been chosen to assess the consistency between the recorded and predicted flow rates: the root-mean square error (RMSE) and the coefficient of determination. The two models showed comparable performance. In particular, both models can properly model the hydrograph shape, the time to peak and the total runoff. The SVR algorithm tends to underestimate the peak discharge, while SWMM tends to overestimate it. SVR shows great potential for applications in the field of urban hydrology, but currently it also has significant limitations regarding the model calibration.

Description: Peak water demand is one of the most stringent operative conditions for a Water Distribution System (WDS), not only for the intensity of the event itself, but also for its recurring nature. The estimation of the maximum water demand is a crucial aspect in both the design and management processes. Studies in the past have tackled this issue with deterministic approaches, even if peak phenomena are distinctly random. In this work, probabilistic models have been developed to study and forecast the daily maximum residential water demand. Some probability distributions have been tested by means of statistical inferences on different data samples related to three monitored WDS. The parameter estimations of the proposed equations have been related to the number of supplied users. Furthermore, this work investigates time scaling effects on the effectiveness of the proposed distributions and relations. Corrective factors that take into account the effect of time averaging step on the above-mentioned parameters have been proposed.

Description: Water, Vol. 10, Pages 629: Experimental Analysis of the Hydraulic Performance of Filtering Cartridges in Drinking Water Networks Water doi: 10.3390/w10050629 Authors: Giacomo Viccione Stefania Evangelista Giovanni de Marinis Liquid treatment processes have been assuming increasing importance in recent decades with the progressive industrialization to ensure public health security for drinking water or to prevent economic damage when safeguarding important production processes. Major investments have been devoted to the research, study, and design of innovative products that are able to respond to the demands of the market, which currently offer several solutions, among which filtration treatment still represents a major one. This work focuses, in particular, on filtration of drinking water with filter cartridges, with the aim to test their hydraulic performance and, particularly, to evaluate the head losses that they produce when introduced into a hydraulic system. The local pressure drops, in fact, may compromise hydraulic plants already characterized by low pressures. What is more, this condition is increasingly likely in supplying networks due to the coexistence of several factors, such as water losses due to failures and inefficient maintenance, severe and prolonged droughts, and increased water demand related to social and economic development. In these systems, the insertion of filtration cartridges can make the pressure levels fall below the minimum limit recommended to ensure the smooth operation of domestic devices. More in detail, in the present study the behavior of seven different commercial filter cartridges was analyzed through a set of experiments conducted in a pilot circuit at the Laboratory of Environmental and Maritime Hydraulics (LIDAM), University of Salerno. These tests have been performed in different operating conditions, collecting pressure data through various pressure gauges. The analysis proved that for common values of operating flow rates in domestic plants the pressure drops in the filter can be of the order of some meters and provided some useful information for the choice of the proper cartridge in low-pressure distribution systems.

Description: Stormwater runoff is often contaminated by human activities. Stormwater discharge into  water bodies significantly contributes to environmental pollution. The choice of suitable treatment  technologies is dependent on the pollutant concentrations. Wastewater quality indicators such as  biochemical oxygen demand (BOD5), chemical oxygen demand (COD), total suspended solids (TSS),  and total dissolved solids (TDS) give a measure of the main pollutants. The aim of this study is to  provide an indirect methodology for the estimation of the main wastewater quality indicators, based  on some characteristics of the drainage basin. The catchment is seen as a black box: the physical  processes of accumulation, washing, and transport of pollutants are not mathematically described.  Two models deriving from studies on artificial intelligence have been used in this research: Support  Vector Regression (SVR) and Regression Trees (RT). Both the models showed robustness, reliability,  and high generalization capability. However, with reference to coefficient of determination R2 and  root‐mean square error, Support Vector Regression showed a better performance than Regression  Tree in predicting TSS, TDS, and COD. As regards BOD5, the two models showed a comparable  performance. Therefore, the considered machine learning algorithms may be useful for providing  an estimation of the values to be considered for the sizing of the treatment units in absence of direct  measures.

Description: Water, Vol. 10, Pages 309: Experimental Analysis of the Hydraulic Performance of Wire-Wound Filter Cartridges in Domestic Plants Water doi: 10.3390/w10030309 Authors: Giacomo Viccione Stefania Evangelista Giovanni de Marinis Among the treatment processes in water networks—of increasing importance in recent decades due to the progressive deterioration of water quality—filtration still represents a major solution. The present work focuses in particular on the filtration of drinking water with wire-wound filter cartridges, the most widely used type of cartridge in domestic plants among the commercially available cartridges, due to their efficiency and relatively low costs. Specifically, the hydraulic performance of these cartridges was analyzed, i.e., mainly the effect of their introduction into a hydraulic system in terms of head losses. The local pressure drops produced by the cartridges may, in fact, create problems in hydraulic plants already characterized by low pressures, where pressure levels may fall below the minimum limit recommended to ensure the smooth operation of domestic devices. To this aim, a set of experiments was conducted in a pilot circuit in the Laboratory of Environmental and Maritime Hydraulics (LIDAM) at University of Salerno, where pressure drops produced by the cartridges were measured in different operating conditions. The artificially dirty conditions of the wire-wound filters were analyzed in order to evaluate the effect of the filter obstruction. The analysis provided some useful information about the performance and duration of these filters, as well as suggestions for more efficient commercial filters.

Description: Side weirs have been widely used since ancient times in many hydraulic works. Their operation can be analyzed following different approaches. However, almost all possible analysis approaches require knowledge of the discharge coefficient, which depends on several geometric and hydraulic parameters. An effective methodology for predicting discharge coefficient can be based on machine learning algorithms. In this research, experimental data obtained from tests carried out on a side weir in a circular channel and supercritical flow have been used to build predictive models of the equivalent discharge coefficient, by which the lateral outflow can be estimated by referring only to the flow depth upstream of the side weir. Four models, different in the input variables, have been developed. Each model has been proposed in 5 variants, depending on the applied algorithm. The focus is mainly on two lazy machine learning algorithms: k Nearest Neighbor and K-Star. The 5-input variables Model 1 and the 4-input variables Model 2 noticeably outperform the 3-input variables Model 3 and Model 4, showing that a suitable characterization of the side weir geometry is essential for a good accuracy of the prediction model. In addition, under models 1 and 2, k Nearest Neighbor and K-Star, despite the simpler structure, provide comparable or better performance than more complex algorithms such as Random Forest and Support Vector Regression.