Bridge afflux analysis through arched bridge constrictions using artificial intelligence methods

Although many studies have been carried out for estimating the afflux through modern straight deck bridge constrictions, little attention has been given to medieval arched bridge constrictions. Hydraulic Research Wallingford in the UK (Brown, P.M., 1988 Brown, P. M. 1988. Afflux at arch bridges, Wallingford, , UK: HR Wallingford. Report SR 182 [Google Scholar]. Afflux at arch bridges. Report SR 182. Wallingford, UK: HR Wallingford) recently published a major coverage of both experimental and field afflux data obtained from arched bridge constrictions. The report pointed out that the present day formulas developed for estimating the bridge afflux are inadequate to apply to ancient arched structures. Therefore, this study aimed at developing new afflux methods for arched bridge constrictions using multi-layer perceptrons (MLP) neural networks, radial basis function-based neural networks (RBNN), generalised regression neural networks (GRNN) and adaptive neuro-fuzzy inference system (ANFIS) model. Multiple linear and multiple nonlinear regression analyses were also used for comparison purposes. Mean square errors, mean absolute errors, mean absolute relative errors, average of individual ratios between predicted and actual values, and determination coefficients were used as comparison criteria for the evaluation of model performances. The test results showed that MLP, RBNN, GRNN, and ANFIS models gave reasonable accuracy when applied to both the field and experimental data collected by Hydraulic Research Wallingford.     

Evapotranspiration Modeling Using Linear Genetic Programming Technique

The study investigates the accuracy of linear genetic programming (LGP), which is an extension to genetic programming (GP) technique, in daily reference evapotranspiration (ET0) modeling. The daily climatic data, solar radiation, air temperature, relative humidity, and wind speed from three stations, Windsor, Oakville, and Santa Rosa, in central California, are used as inputs to the LGP to estimate ET0 obtained using the FAO-56 Penman-Monteith equation. The accuracy of the LGP is compared with those of the support vector regression (SVR), artificial neural network (ANN), and those of the following empirical models: the California irrigation management system Penman, Hargreaves, Ritchie, and Turc methods. The root-mean-square errors, mean-absolute errors, and determination coefficient (R2) statistics are used for evaluating the accuracy of the models. Based on the comparison results, the LGP is found to be superior alternative to the SVR and ANN techniques.     

Prediction of Hydropower Energy Using ANN for the Feasibility of Hydropower Plant Installation to an Existing Irrigation Dam

Recently, artificial neural networks (ANNs) have been used successfully for many engineering problems. This paper presents a practical way of predicting the hydropower energy potential using ANNs for the feasibility of adding a hydropower plant unit to an existing irrigation dam. Because the cost of energy has risen considerably in recent decades, addition of a suitable capacity hydropower plant (HPP) to the end of the pressure conduit of an existing irrigation dam may become economically feasible. First, a computer program to realistically calculate all local, frictional, and total head losses (THL) throughout any pressure conduit in detail is coded, whose end-product enables determination of the C coefficient of the highly significant model for total losses as: THL = C·Q ². Next, a computer program to determine the hydroelectric energies produced at monthly periods, the present worth (PW) of their monetary gains, and the annual average energy by a HPP is coded, which utilizes this simple but precise model for quantification of total energy losses from the inlet to the turbine. Inflows series, irrigation water requirements, evaporation rates, turbine running time ratios, and the C coefficient are the input data of this program. This model is applied to randomly chosen 10 irrigation dams in Turkey, and the selected input variables are gross head and reservoir capacity of the dams, recorded monthly inflows and irrigation releases for the prediction of hydropower energy. A single hidden-layered feed forward neural network using Levenberg–Marquardt algorithm is developed with a detailed analysis of model design of those factors affecting successful implementation of the model, which provides for a realistic prediction of the annual average hydroelectric energy from an irrigation dam in a quick-cut manner without the excessive operation studies needed conventionally. Estimation of the average annual energy with the help of this model should be useful for reconnaissance studies.     

Microcavity Formation in Alumina Using Ti Templates II: Mechanism and Kinetics

In part I of this work, it was found that titanium (Ti) wire encapsulated within mechanically milled alumina powder and sintered at 1350°C forms potentially useful microcavities due to the consolidation of Kirkendall porosity. Here a series of samples sintered at 1350°C in the range 0–24 h has shown the remarkable way in which these cavities form. The cavity has already started in samples quenched from the top of the heating ramp (0 min at 1350°C). It is surrounded by a diffusion zone ∼300 μm in diameter, which does not change size throughout the firing process although the contents change markedly. The diffusion zone microstructure is initially complex with phase sequence TiO₂/Al₂O₃/TiO₂+Al₂O₃/Al₂TiO₅. Microstructure evolution may be summarized as outward growth of the cavity accompanied by inward growth of the Al₂TiO₅ resulting in a ∼190-μm-diameter cavity surrounded by a 50-μm-thick layer of Al₂TiO₅. The formation of the cavity and surrounding microstructure is discussed although some features, such as the nucleation of Al₂TiO₅ in the part of the diffusion zone furthest from the Ti source and the ring of Al₂O₃, which persists in between Ti-rich parts of the diffusion zone are still poorly understood.     

Predicting river daily flow using wavelet-artificial neural networks based on regression analyses in comparison with artificial neural networks and support vector machine models

This study investigates the ability of wavelet-artificial neural networks (WANN) for the prediction of short-term daily river flow. The WANN model is improved by conjunction of two methods, discrete wavelet transform and artificial neural networks (ANN) based on regression analyses, respectively. The proposed WANN models are applied to the daily flow data of Vanyar station, on the Ajichai River in the northwest region of Iran, and compared with the ANN and support vector machine (SVM) techniques. Mean square error (MSE), mean absolute error (MAE) and correlation coefficient (R) statistics are used for evaluating precision of the WANN, ANN and SVM models. Comparison results demonstrate that the WANN model performs better than the ANN and SVM models in short-term (1-, 2- and 3-day ahead) daily river flow prediction.

     

Intermediate Phases in Ti3SiC2 Synthesis from Ti/SiC/C Mixtures Studied by Time-Resolved Neutron Diffraction

The reactive sintering of 3Ti/SiC/C to form the layered ternary carbide Ti₃SiC₂ was studied in situ by time-resolved neutron powder diffraction. A number of intermediate processes occur during the synthesis beginning with the α-β transition in Ti. Concurrent with the α-β transition, two intermediate phases, TiC _x and Ti₅Si₃C _x ( x ≤ 1), form. These phases account for almost the entire sample in the range 1500–1600°C beyond which they react with each other and a small amount of free C to form the product phase Ti₃SiC₂.     

In Situ Neutron Diffraction Study of a Liquid Nitrogen-Quenched Mg-PSZ Under Load: A Microcrack-Dominated System?

The mechanical response of liquid nitrogen-quenched 9.4 Mg-PSZ in which the orthorhombic ( o ) phase is the major constituent (46 wt%) was investigated using in-situ neutron diffraction during uniaxial compression. The material remains elastic below 1 GPa with a Young's modulus of∼242 GPa, second highest of all zirconia-based materials and highest of all zirconia-based ceramics. Beyond 1 GPa, the material develops small plastic strains in a time-dependent manner (i.e., by room temperature creep) although the strains were generally much smaller than the unquenched material, which contains no o phase. As for standard Mg-PSZ, the creep was accompanied by a volume change usually indicative of tetragonal to monoclinic ( m ) phase transformation; however, the amount of m phase apparent in the neutron diffraction patterns increased only marginally. The magnitude of the volume increase could not be accounted for by the observed increase in the m phase and hence, microcracking is believed to be responsible for most of the volume change. There is some evidence for a small amount of o to m transformation at the detection limit of the phase analysis technique.     

Crystal Structure of Orthorhombic Zirconia in Partially Stabilized Zirconia

A neutron powder diffraction investigation confirms that, in tough magnesiapartially-stabilized zirconia cooled to 30 K, most of the tetragonal zirconia transforms to an orthorhombic phase. This phase is retained on heating to room temperature; the lattice parameters at 295 K are a = 0.5068, b = 0.5260, and c = 0.5077 nm. The room-temperature crystal structure (space group Pbc2₁) is determined by multiphase Rietveld refinement from the neutron diffraction pattern. This orthorhombic structure is compared with the parent tetragonal structure and with the structure of monoclinic zirconia, which it closely resembles.