Solving the stochastic support vector regression with probabilistic constraints by a high-performance neural network model

This paper offers a recurrent neural network to support vector machine (SVM) learning in stochastic support vector regression with probabilistic constraints. The SVM is first converted into an equivalent quadratic programming (QP) formulation in linear and nonlinear cases. An artificial neural network for SVM learning is then proposed. The presented neural network framework guarantees obtaining the optimal solution of the SVM problem. The existence and convergence of the trajectories of the network are studied. The Lyapunov stability for the considered neural network is also shown. The efficiency of the proposed method is shown by three illustrative examples.

     

Enhanced oil recovery from carbonate reservoirs by spontaneous imbibition of low salinity water

An experimental study was performed to investigate the impact of low salinity water on wettability alteration in carbonate core samples from southern Iranian reservoirs by spontaneous imbibition. In this paper, the effect of temperature, salinity, permeability and connate water were investigated by comparing the produced hydrocarbon curves. Contact angle measurements were taken to confirm the alteration of surface wettability of porous media. Oil recovery was enhanced by increasing the dilution ratio of sea water, and there existed an optimum dilution ratio at which the highest oil recovery was achieved. In addition, temperature had a very significant impact on oil recovery from carbonate rocks. Furthermore, oil recovery from a spontaneous imbibition process was directly proportional to the permeability of the core samples. The presence of connate water saturation inside the porous media facilitated oil production significantly. Also, the oil recovery from porous media was highly dependent on ion repulsion/attraction activity of the rock surface which directly impacts on the wettability conditions. Finally, the highest ion attraction percentage was measured for sodium while there was no significant change in pH for all experiments.     

Representation of Dynamic Positioning on Hovering Type Autonomous Underwater Vehicle in Precise Welding Operations

This paper proposes dynamic positioning (DP) on a hovering autonomous underwater vehicle (HAUV) to perform accurate underwater processes such as welding operations. High maneuverability, high controllability, and hovering of the robot were the prerequisites of this operation, which increase its accuracy and velocity and reduce costs and human health risks. Other types of thrusters were used in this robot to reduce the number of thrusters and controller''s complexity. Controlling every 6 degrees of freedom to perform this type of operation was done. Furthermore, such a delicate operation required controlling the translational and rotational movements together. There was also a need to control the velocities to travel in a prescribed distance at a reasonable time. The possibility of dynamic positioning for welding and maintaining position at a point was defined for the robot. Then the robot''s performance under a defective state-servo motor failure, thruster malfunction-and the subsequent effects on the performance during the predetermined missions were investigated. Simulation results demonstrated that the HAUV has the capability to perform dynamic positioning operations. In this article, one of the prevalent classic control methods called PID controller was employed for controlling the movements of the robot.     

Comparison of interwell connectivity predictions using percolation, geometrical, and Monte Carlo models

Reservoir connectivity is often an important consideration for reservoir management. For example, connectivity controls waterflood sweep efficiency and it affects decisions concerning well placement and spacing. The uncertainty of sandbody distributions, however, can make interwell connectivity prediction extremely difficult. Percolation models are a useful tool to simulate sandbody connectivity behavior to estimate interwell connectivity.This study applies a percolation method to estimate interwell connectivity. Using results derived by Andrade, King, and others for fluid travel time between locations in a percolation model, we develop a method to estimate interwell connectivity. Four parameters are needed to use this approach: the net-to-gross ratio p_sand, the typical sandbody size, reservoir length and well spacing. To evaluate this new percolation method, the results are compared to results from geometrical models, Monte Carlo, and reservoir simulation.These methods were applied to estimate interwell connectivity for three non-communicating stratigraphic intervals in Monument Butte oil field, Utah. The results suggest that the percolation method can estimate the probability of interwell connectivity reliably for thin intervals for any values of p_sand, well spacing, and reservoir length. The geometrical model also performs well, but can only be applied in fields where the well spacing is less than one-half of the sandbody size.The proposed method requires that the reservoir interval for evaluation be sufficiently thin so that 2D percolation results can be applied. For thick intervals or heterogeneous sandbody distributions, the percolation method developed here is not suitable because it assumes thin layers. Future percolation research will be needed to adapt this new method to 3D cases.     

Equilibrium moisture content and heat of desorption of saffron

The equilibrium moisture contents of saffron (Crocus sativus L.) stigmas were determined experimentally using the standard gravimetric method at temperatures 30, 45 and 60 °C and water activity ranging from 11% to 83%. The sorption isotherm curves of saffron were sigmoidal in shape and decreased with increased temperature at constant relative humidity. Five selected isotherm models GAB, modified Henderson, modified Chung‐Pfost, modified Halsaey and modified Oswin were tested to fit the experimental isotherm data. Modified Oswin and modified Henderson models were found acceptable for predicting desorption moisture isotherms and fitting to the experimental data, respectively. The isosteric heats of desorption, determined from equilibrium data using the Clausius‐Clapeyron equation, were found to be a function of moisture content. The net isosteric heat of desorption of saffron varied between 1.38 and 5.38 kJ mol^?1 at moisture content varying between 2% and 20% (d.b).     

Microstructural and mechanical assessment of transient liquid phase bonded commercially pure titanium

Diffusion joining of commercially pure titanium was successfully prepared via transient liquid phase bonding in vacuum environment. The process was carried out using AMS 4772 silver-based filler alloy at 900–1000°C for various holding time under the vacuum of 6?×?10^?7?Torr. Optical and scanning electron microscopy equipped with an EDS analyzer was conducted for microstructural evaluations. Mechanical properties were also investigated by shear test, fractographic assessment and X-ray diffraction analyses. The tendency to achieve isothermally solidified joint increased by increasing bonding time. No sign of athermal solidification was detected of sample bonded at 1000°C for 90?min. Consequently, the bonding condition of a high quality joint was obtained. Elemental analyses revealed that filler alloy’s elements (Ag, Cu) distributed more uniformly in fully isothermal solidified bond, whereas the aggregation of these elements is considerable in athermally solidified bond. Shear test results represented that the highest shear strength attributed to the sample bonded in isothermal solidified condition (bonded at 1000°C for 90?min).     

Lumped Parameter Thermal Model for Axial Flux Switched Reluctance Motors

The accurate prediction of temperature rise in various parts of a machine plays a significant role in ensuring that the machine delivers the required performance avoiding failures that arise due to high temperatures. In this paper, a lumped parameter thermal model is presented for exterior-rotor axial flux switched reluctance motors. Depending on its geometry, each part of the motor is modeled as a thermal equivalent circuit based on general cylindrical or cuboidal components. A total thermal network is obtained by proper connection of these sub-circuits. The model was calibrated by a stationary test; then, running conditions were applied to the model. The sensitivity of the model to the heat transfer coefficients was investigated by considering the effect of the coefficient variation on the estimated steady-state temperatures. Results show that the winding temperature, as the hottest part of the motor, is the most sensitive to the frame-to-ambient heat transfer coefficients.     

Vehicle mass and injury risk in two-car crashes: A novel methodology

This paper introduces a novel methodology based on disaggregate analysis of two-car crash data to estimate the partial effects of mass, through the velocity change, on absolute driver injury risk in each of the vehicles involved in the crash when absolute injury risk is defined as the probability of injury when the vehicle is involved in a two-car crash. The novel aspect of the introduced methodology is in providing a solution to the issue of lack of data on the speed of vehicles prior to the crash, which is required to calculate the velocity change, as well as a solution to the issue of lack of information on non-injury two-car crashes in national accident data. These issues have often led to focussing on relative measures of injury risk that are not independent of risk in the colliding cars. Furthermore, the introduced methodology is used to investigate whether there is any effect of vehicle size above and beyond that of mass ratio, and whether there are any effects associated with the gender and age of the drivers. The methodology was used to analyse two-car crashes to investigate the partial effects of vehicle mass and size on absolute driver injury risk. The results confirmed that in a two-car collision, vehicle mass has a protective effect on its own driver injury risk and an aggressive effect on the driver injury risk of the colliding vehicle. The results also confirmed that there is a protective effect of vehicle size above and beyond that of vehicle mass for frontal and front to side collisions.