Bottom-Up Generation of Peak Demand Scenarios in Water Distribution Networks

Creaco, Enrico; Galuppini, Giacomo; Campisano, Alberto; Franchini, Marco

doi:10.3390/su13010031

Cited by 12 publications

(12 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For N = 100 there is some skewness, which is consistent with the findings of Creaco et al. (2021), but data starts to resemble a normal probability plot. Data fits a normal probability plot for the total population ( N = 79,106).…”

Section: Resultssupporting

confidence: 90%

“…This is because averaged water demands are predominantly zero, and so a normal distribution cannot be assumed for water demands. For N = 100 there is some skewness, which is consistent with the findings of Creaco et al (2021), but data starts to resemble a normal probability plot. Data fits a normal probability plot for the total population (N = 79,106).…”

Section: Validation Case Studysupporting

confidence: 84%

“…However, when a sufficient amount of population is assessed, the aggregated water consumption can be assumed to follow a normal distribution with mean μ Q and standard deviation σ Q . Other authors have proved that skewness is non negligible for low aggregation levels based on measurement series (Creaco et al, 2021), but the normal assumption is here adopted as a conceptual simplification and will be validated in the Results section. The mean and standard deviation statistical properties are not constant: they vary with the spatial aggregation level being considered (N number of inhabitants) and they depend on the temporal framework adopted for demand analysis.…”

Section: Peak Demand Valuesmentioning

confidence: 99%

“…Creaco et al. (2021) have also recently worked in developing a two‐step methodology for the generation of snapshot peak demand scenarios, each of which is based on a single combination of demand values at nodes.…”

Section: Introductionmentioning

confidence: 99%

“…Del Giudice et al ( 2020) go one step further and propose a methodological framework to estimate the expected value of hourly peak demand factors considering its dependence on the spatial aggregation level. Creaco et al (2021) have also recently worked in developing a two-step methodology for the generation of snapshot peak demand scenarios, each of which is based on a single combination of demand values at nodes.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Short‐Term Variability Effect on Peak Demand: Assessment Based on a Microcomponent Stochastic Demand Model

Díaz

González

2022

Water Resources Research

View full text Add to dashboard Cite

Peak demand has an important role in water distribution system design because it is associated with one of the most burdensome operating scenarios in a network. For this reason, peak demand has been widely discussed over the last century. Traditionally, peak demand assessment has focused on determining the peak demand coefficient or peak factor, which is defined as the ratio between the maximum and the mean daily flow. Authors like Harmon (1918), Babbitt (1928, Metcalf and Eddy (1935), or Johnson (1942) (among others) proposed expressions that provide the peak factor at sewer systems based on the size of the population. These empirical equations set up the basis for peak demand analysis at water supply systems, although they address the problem from a deterministic point of view.Water demand is nowadays recognized as one of the main random factors that condition flow variability (Magini et al., 2008). The development of stochastic demand models that simulate the complex pulsed nature of water demands has motivated the shift toward probabilistic (rather than deterministic) demand analysis (Vertommen et al., 2015), also for peak demand assessment. According to the literature review presented by Creaco, Blokker, and Buchberger (2017), stochastic demand models can be broadly classified as: (a) household-based models, which adjust statistical models based on flow measurements at monitored households, like the Poisson Rectangular Pulse (PRP) model originally presented by Buchberger and Wu (1995), and (b) end-use models, which compute household consumption by aggregating the contribution of each end-use or microcomponent (e.g., taps, showers, washing machines) according to survey-based data. Zhang et al. (2005) and Creaco et al. ( 2018) have already applied PRP-like models to assess peak demands. Zhang et al. (2005) proposed a theoretical explanation for peaking factors by combining a PRP model with extreme value theory. Balacco et al. (2017) adapted this approach to a case study in Italy, comparing it with traditional formulas and real measurements. On the other hand, SIMDEUM (SIMulation of water Demands, an End-Use Model) is the reference microcomponent model at present (Creaco, Blokker, & Buchberger, 2017), and it has also been used to compute accurate estimates of peak demands (Blokker et al., 2012).Addressing peak demand assessment probabilistically is not the only current challenge. The temporal and spatial resolution effect on peak factors is being discussed as well. As pointed out by Tricarico et al. (2007), assuming a time interval of one hour (traditional temporal framework for peak assessment) may result in an underestimation of peak demand, because major peaks could take place within the hour. On the contrary, using very fine time scales (e.g., 1 s) is excessive considering that per second variations are not expected to be decisive in order

show abstract

Section: Resultssupporting

confidence: 90%

Section: Validation Case Studysupporting

confidence: 84%

Section: Peak Demand Valuesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations