Dynamic time warping improves sewer flow monitoring

Correlation and spectral analyses to assess the response of a shallow aquifer to low and high frequency rainfall fluctuations

“…An approach of this kind was first proposed by Beck et al (1969) to determine flow velocity for manufacturing purposes. Latest applications in geosciences include the estimation of debris flow rate (Arattano and Marchi, 2005), assessment of transit time in karst systems (Bailly-Comte et al, 2008 and sewer flow monitoring (Dürrenmatt et al, 2013). Delbart et al (2014) used time-varying cross-correlation functions to define the dynamic response time of shallow karstic GWLs to precipitation input.…”

Mixed dissimilarity measure for piecewise linear approximation based time series applications

“…Due to the expected better results and the speed up, most practitioners dealing with DTW utilizing of some form of global constraints independently of whether DTW was used for off-line handwritten signature verification (Güler & Meghdadi, 2008) or for sewer flow monitoring (Dürrenmatt et al, 2013). However, selecting the optimal global constraint -deciding not only which constraint should be used, but whether any constraint should be used at all -is not obvious.…”

“…Such methods have rapidly gained attention in engineering applications, including process engineering (Singhal & Seborg, 2005), medicine (Tormene, Giorgino, Quaglini, & Stefanelli, 2009), bioinformatics (Aach & Church, 2001), chemistry (Abonyi, Feil, Németh, & Árva, 2005), finance (Rada, 2008), biometrics (Gavrila & Davis, 1995;Kholmatov & Yanikoglu, 2004;Vajna, 2000) and even tornado prediction (McGovern, Rosendahl, Brown, & Droegemeier, 2011) and sewer flow monitoring (Dürrenmatt, Giudice, & Rieckermann, 2013), because of the effective and efficient methods available for solving time series-related problems. But these methods are not limited exclusively to temporal data.…”

Rhodamine WT tracer experiments to check flow measurements in sewers

“…The first simple one based only on three points according to the manufacturer protocol [10] is indeed not sufficient to assess uncertainties in measurements and calibration function. Therefore a second step calibration is performed, using a set of N ¼14 (in this study) Rhodamine WT standard solutions with concentration c sV ranging from 0 to 130 g/m 3 , with a step of 10 g/m 3 5)) (see Fig. 3), by applying the Williamson regression method which minimizes the sum S as given by Eq.…”

“…For such cases, discharge values provided by sensors need to be carefully checked by independent methods like tracer experiments, exploration of the velocity field with current meter or portable electromagnetic flow meter or CFD modeling of velocity fields e.g. [1][2][3].…”