A key physical property used in the description of a soil-water regime is a soil water retention curve, which shows the relationship between the water content and the water potential of the soil. Pedotransfer functions are based on the supposed dependence of the soil water content on the available soil characteristics, e.g., on the relative content of the particle size in the soil and the dry bulk density of the soil. This dependence could be extracted from the available data by various regression methods. In this paper, artificial neural networks (ANNs) and support vector machines (SVMs) were used to estimate a drying branch of a water retention curve. The paper compares the mentioned methods by estimating the water retention curves on regional scale for the Záhorská lowland in the Slovak Republic, where relatively small data set was available. The performance of the models was evaluated and compared. These computations did not fully confirm the superiority of SVMs over ANNs as is often proclaimed in the literature, because the results obtained show that in this study the ANN model performs somewhat better and is easier to handle in determining pedotransfer functions than the SVM models. Nevertheless, the results from both data-driven models are quite close, and the results show that they provide a significantly more precise outcome than a traditional multi-linear regression does., Autori sa v príspevku venujú určovaniu pedotransferových funkcií (PTF), ktoré umožňujú stanoviť body vlhkostných retenčných kriviek pôdy z ľahšie merateľných pôdnych vlastností a sú dôležitým prvkom modelovania vodného režimu pôdy. Ešte v minulej dekáde sa objavili snahy využívať na ich určenie umelé neurónové siete (UNS). Multi-layer perceptron (MLP) čiže viacvrstvový perceptrón je najčastejšie používaný model doprednej umelej neurónovej siete s kontrolovaným typom učenia. Vstupné signály prechádzajú sieťou typu MLP iba dopredným smerom, teda postupne od vrstvy k vrstve. MLP používa tri a viac vrstiev neurónov rozdelených na vstupnú, skrytú a výstupnú vrstvu s nelineárnou aktivačnou funkciou a vie rozpoznať alebo modelovať informácie, ktoré nie sú lineárne oddeliteľné alebo závislé. Novší vývoj v oblasti učiacich algoritmov poskytuje ďalšie možnosti, z ktorých sa v tomto príspevku venujeme tzv. mechanizmom podporných vektorov (Support Vector Machines - SVM). SVM využíva pri svojom kalibrovaní na riešený problém princíp tzv. štrukturálnej minimalizácie namiesto iba minimalizácie chyby - (Vapnik, 1995). Pri trénovaní siete MLP je jediným cieľom minimalizovať celkovú chybu. Pri SVM sa simultánne minimalizuje chyba aj zložitosť modelu. Použitie tohto princípu vedie zvyčajne k vyššej schopnosti generalizácie, t.j. umožneniu presnejších predpovedí pre dáta, ktoré neboli použité pri trénovaní SVM. Vhodnosť štandardnej umelej neurónovej siete, SVM a viacnásobnej lineárnej regresie sa v článku vyhodnocuje na základe údajov získaných z pôdnych vzoriek odobratých v lokalite Záhorskej nížiny. Pôvodné údaje a ich aplikáciu pri vyhodnocovaní vodného režimu pôd uvádza Skalová (2001, 2007), odkiaľ boli prevzaté vstupné dáta a to percentuálny obsah zrnitostných kategórií (I až IV podľa Kopeckého), redukovaná objemová hmotnosť (ρd) a vlhkosti pre vlkostné potenciály hw= -2.5, -56, -209, -558, -976, -3060, -15300 cm, ktoré boli stanovené laboratórne pre potreby určenia a testovania regresných závislostí. Vzhľadom na to, že pri odvodzovaní regionálnych PTF je častým prípadom nedostatok dát pre odvodenie dátovo riadených modelov, autori navrhli riešiť úlohu pomocou ansámblu MLP resp. SVM. Ansámbel dátovo riadených modelov bol vytvorený variabilným rozdelením údajov na trénovacie a validačné (validačnými údajmi sa testuje presnosť modelu vo fáze jeho tvorby, ešte sa používajú konečné testovacie dáta, ktoré neboli pri tvorbe modelu použité). Výsledky ukázali lepšie regresné schopnosti oboch dátovo riadených modelov (SVM aj MLP) voči multilineárnej regresii a o niečo lepšie výsledky boli získané z viacvrstvového perceptrónu než zo SVM., and Keďže v niektorých iných prácach mal zvyčajne vyššiu výpočtovú presnosť model založený na SVM než na UNS, autori odporúčajú pre budúci výskum preveriť vhodnosť kombinácie SVM a MLP modelov v dátovo riadenom skupinovom modeli.
The erosion, transport and deposition of sediments in small valley reservoirs represent a significant impact on their operations, mainly with regard to reducing the volume of their accumulation. The aim of this study is a comparison and uncertainty analysis of two modelling concepts for assessment of soil loss and sediment transport in a small agricultural catchment, with an emphasis on estimating the off-site effects of soil erosion resulted in sedimentation of a small water reservoir. The small water reservoir (polder) of Svacenicky Creek which was built in 2012, is a part of the flood protection measures in Turá Lúka and is located in the western part of Slovakia, close to the town of Myjava. The town of Myjava in recent years has been threatened by frequent floods, which have caused heavy material losses and significantly limited the quality of life of the local residents. To estimate the amount of soil loss and sediments transported from the basin, we applied two modelling concepts based on the USLE/SDR and WaTEM/SEDEM erosion models and validated the results with the actual bathymetry of the polder. The measurements were provided by a modern Autonomous Underwater Vehicle (AUV) hydrographic instrument. From the sediment data measured and the original geodetic survey of the terrain conducted at the time of the construction of the polder, we calculated changes in the storage volume of the polder during its four years of operation. The results show that in the given area, there has been a gradual clogging of the bottom of the polder caused by water erosion. We estimate that within the four years of the acceptance run, 10,494 m3 of bottom sediments on the Svacenicky Creek polder have accumulated. It therefore follows that repeated surveying of the sedimentation is very important for the management of the water reservoir.
The heterogeneity of water flow and solute transport was assessed during radioactive tracer infiltration experiment in a black clay loam soil using modified methods to estimate the effective cross section (ECS) and the degree of preferential flow (DPF). The results of field and numerical experiments showed that these parameters characterized the heterogeneity of water flow in the soils unequivocally. The ECS decreases non-linearly and the DPF increases linearly with an increase of the bypassing ratio (ratio of macropore flow rate to total flow rate). The ECS decreased and the DPF increased with depth, which suggests an increase in the heterogeneity of water flow with depth. The plot of the DPF against ECS values calculated from the tracer experiment data was consistent with the relationship obtained by the numerical simulation assuming preferential flow in the neighbourhood of three probes.
The paper describes qualitative analysis of contaminant transport in a homogeneous, isotropic aquifer where first order chemical reaction at the boundary and nonlinear decay act simultaneously. In order to analyze the relative merits, two types of disposal scheme are considered; (i) Scenario I: longer duration with low input concentration and (ii) Scenario II: short duration with higher input concentration. The governing advective-dispersion equation is solved numerically by employing ADI scheme under finite difference method. We apply the method of moments to investigate mean concentration distribution and other statistical parameters such as central moment, coefficients of skewness (β2 ) and kurtosis (β 3 ). The mean concentration distribution ( Cm ) is computed by applying Edgeworth’s asymptotic series for non-Gaussian curves involving Hermite polynomials ( Hn ). The forward displacement of centroid ( Xg ) with time, deviations of mean concentration distribution from Gaussianity and breakthrough curves have been examined. and Príspevok obsahuje kvalitatívnu analýzu transportu kontaminantov v homogénnej, izotropnej zvodni, na hraniciach ktorej simultánne prebiehajú chemické reakcie prvého rádu a nelineárny rozpad. Aby sme mohli posúdiť relatívne výhody spôsobu analýzy, použili sme dva typy schém; (i) Scenár I: nízka koncentrácia vstupov a ich dlhšie trvanie; (ii) Scenár II: krátko trvajúce vysoké koncentrácie vstupov. Advektívnodisperzná rovnica je riešená numericky, s využitím schémy ADI v rámci metódy konečných rozdielov. Na určovanie rozdelenia priemerných koncentrácií a iných štatistických parametrov, ako je centrálny moment, koeficient šikmosti (β2 ) a strmosť (β 3 ) , použili sme metódu momentov. Rozdelenie priemerných koncentrácií ( Cm ) sme vypočítali aplikáciou Edgeworthových asymptotických radov negaussovských kriviek, obsahujúcich Hermitove polynómy ( Hn ). Študovali sme dopredný posun centroidu ( Xg ) v závislosti od času, odchýlky od priemerných hodnôt rozdelenia priemerných koncentrácií od gaussovských, a určili sme tiež prienikové krivky.
Carbonate rocks host several large water and hydrocarbon reservoirs worldwide, some of them highly heterogeneous involving complex pore systems. Pre-salt reservoirs in the Santos Basin off the south-east coast of Brazil, are an example of such rocks, with much attention focused on proper characterization of their petrophysical and multiphase flow properties. Since it is very difficult to obtain rock samples (coquinas) from these very deep reservoirs, analogues from north-eastern Brazil are often used because of very similar geological age and petrophysical properties. We used a coquina plug from an outcrop in a quarry in northeast Brazil to perform a comprehensive set of analyses. They included Scanning Electron Microscopy (SEM), Energy-Dispersive X-ray Spectroscopy (EDS), X-ray Diffraction (XRD), and micro-computed tomography (μCT) image acquisition using a series of pixel sizes, as well as direct permeability/ porosity measurements. Some of the experimental data were collected from the plug itself, and some from a small sample of the rock slab, including thin sections. Results included the carbonate rock composition and the pore system at different scales, thus allowing us to reconstruct and model the porosity and absolute permeability of the coquina using 3D digital imaging and numerical simulations with pore network models (PNMs). The experimental and numerical data provided critical information about the well-connected pore network of the coquina, thereby facilitating improved predictions of fluid flow through the sample, with as ultimate objective to improve hydrocarbon recovery procedures.
The flow of a mixture of liquid and solid particles at medium and high volume fraction through an expansion in a rectangular duct is considered. In order to improve the modelling of the phenomenon with respect to a previous investigation (Messa and Malavasi, 2013), use is made of a two-fluid model specifically derived for dense flows that we developed and implemented in the PHOENICS code via user-defined subroutines. Due to the lack of experimental data, the two-fluid model was validated in the horizontal pipe case, reporting good agreement with measurements from different authors for fully-suspended flows. A 3D system is simulated in order to account for the effect of side walls. A wider range of the parameters characterizing the mixture (particle size, particle density, and delivered solid volume fraction) is considered. A parametric analysis is performed to investigate the role played by the key physical mechanisms on the development of the two-phase flow for different compositions of the mixture. The main focuses are the distribution of the particles in the system and the pressure recovery.
In this paper a problem of multiple solutions of steady gradually varied flow equation in the form of the ordinary differential energy equation is discussed from the viewpoint of its numerical solution. Using the Lipschitz theorem dealing with the uniqueness of solution of an initial value problem for the ordinary differential equation it was shown that the steady gradually varied flow equation can have more than one solution. This fact implies that the nonlinear algebraic equation approximating the ordinary differential energy equation, which additionally coincides with the wellknown standard step method usually applied for computing of the flow profile, can have variable number of roots. Consequently, more than one alternative solution corresponding to the same initial condition can be provided. Using this property it is possible to compute the water flow profile passing through the critical stage.
Concentration and particle size distribution has been experimentally measured in a 2D rectangular duct under near iso-kinetic conditions for multi-sized particulate slurry. Measurements have been made at different flow velocities for various efflux concentrations in the range of 10 to 50 % by weight. It is observed that the concentration profile is highly skewed towards the bottom of the duct, which reduces with increase in efflux concentration and velocity. Similar phenomenon is observed in the distribution of individual particle size fractions with the effect being more pronounced for the coarser particles. and Rozdělení koncentrace a velikosti částic bylo měřeno ve 2D pravoúhlém kanále při proudění disperze různě velkých částic za téměř iso-kinetických podmínek. Experimenty byly provedeny při různých rychlostech s dopravními koncentracemi v rozsahu 10 až 50 hmotnostních procent. Bylo zjištěno, že koncentrační profil je výrazně zešikmený ke dnu kanálu, což se však zmenšuje s růstem koncentrace a rychlosti. Podobný jev byl pozorován u distribuce částic jednotlivých velikostních frakcí. Jev se projevuje tím výrazněji, čím větší jsou částice.
Sand-water slurry was investigated on an experimental pipe loop of inner diameter D = 100 mm with the horizontal, inclined, and vertical smooth pipe sections. A narrow particle size distribution silica sand of mean diameter 0.87 mm was used. The experimental investigation focused on the effects of pipe inclination, overall slurry concentration, and mean velocity on concentration distribution and deposition limit velocity. The measured concentration profiles showed different degrees of stratification for the positive and negative pipe inclinations. The degree of stratification depended on the pipe inclination and on overall slurry concentration and velocity. The ascending flow was less stratified than the corresponding descending flow, the difference increasing from horizontal flow up to an inclination angle of about +30°. The deposition limit velocity was sensitive to the pipe inclination, reaching higher values in the ascending than in the horizontal pipe. The maximum deposition limit value was reached for an inclination angle of about +25°, and the limit remained practically constant in value, about 1.25 times higher than that in the horizontal pipe. Conversely, in the descending pipe, the deposition limit decreased significantly with the negative slopes and tended to be zero for an inclination angle of about −30°, where no stationary bed was observed.