International Journal of Engineering
Volume:14 Issue: 4, Nov 2001

The non-orthogonal boundary-fitted coordinate transformation method is applied to the solution of steady three-dimensional conservation equations of mass, momentum, energy and speciescontinuity to obtain the laminar velocity, temperature and concentration fields for simulation of polymerization of styrene in arbitrary cross-sectional duct reactors. Variable physical properties (except for specific-heat), viscous heat dissipation and free-convection effects are considered in modeling while axial diffusion is ignored. The conservation equations originally written in Cartesian coordinates are parabolized in the axial direction and then transformed to the non-orthogonal curvilinear coordinates to handle arbitrary duct geometries. The transformed equations are discretized by the control-volume finite-difference approach in which the convective and diffusive terms are handled by the upwind-difference and central-difference approximation schemes respectively. Results are obtained for eight different geometries. The results show that not a specific geometry is in general superior to conventional circular duct reactors for the highest conversion of the chemical reaction under study, considering also the least pressure-drop in the reactors.

(Heavy organics deposition is a common problem in oil industry, especially in oil production, transportation and processing. Wax or solid paraffin series are examples of heavy organics that deposit. Precipitation and crystallization of wax causes major difficulties in different processes. Based on multi-solid theory, a basic model is modified in this paper for wax precipitation in different oils. Four oil samples with paraffin/naphthene/aromatic (PNA) analysis are used. Two parameter gamma function is used for molecular weight distribution of C7+ fraction. Since all samples in original multi-solid model were dead oil with vapor phase being ignored, the pressure and composition effects on wax precipitation were, therefore, impossible (existence of vapor phase). When considering vapor phase in this work, the proposed modification (i.e. solution method) is well able to predict pressure and composition (addition of different solvents) effects on wax appearance point (WAP). The results obtained agree very well with the experimental and previous investigator’s results.

This paper describes the rainfall data generation processes, which were used to simplify the recharge model developed by Khazai and Spink. The principles of techniques used for single and two sites generation are discussed. The application of the techniques for extending the rainfall records at the existing stations and increasing arbitrarily the numbers of rain gauges within the catchment are presented.

This paper describes various time-domain methods useful for analyzing the experimental data obtained from a circular cylinder force in terms of both wave and current for estimation of the drag and inertia coefficients applicable to the Morison’s equation. An additional approach, weighted least squares method is also introduced. A set of data obtained from experiments on heavily roughened circular cylinders in waves and simulated current has been analyzed by all these techniques. The resulting force coefficients are then used to predict the force from separate experiments-results that have not been used in the analysis. The root mean squares error and bias in the estimation of maximum force in each wave cycle is used as a measure of predictive accuracy and as a basis for comparing the analysis techniques. It is found that no single method is consistently better under all circumstances but on average weighted least squares method generally gives the best predictive accuracy by a small margin. The force coefficients obtained by the various methods significantly decrease when current is added to waves.

This paper presents a new control method for simultaneous compensation of instantaneous reactive power and current harmonics by parallel active filters (PAFs). Reference compensating currents of PAF are calculated using the subtraction of instantaneous power from its average value. In this way, it is possible omitting high/low pass filter(s) from the control circuit of PAF, which usually results in phase shift and magnitude change of alternating components of instantaneous power. It is possible regulating the DC side voltage of PAF using a closed loop control circuit, easily. The presented method is compared with the classical p-q theory method through simulation results. A simple model for controlling the switches of power electronic converters is offered that is usable in PSCAD/EMTDC simulation package. Experimental results verify the validity of presented control strategy in generating of reference compensation currents.

The purpose of this paper is to report on a general method, based on time domain samples for spectral manipulation of the time-limited signals. In this direction, all the needed formulas for practical periodic time-limited interpolation in two cases of low pass and band pass have been derived. The work begins with dividing an arbitrary signal in time domain into the time limited nonoverlapping frames. Then each frame is processed to find the effective maximum frequency or equivalently the minimum number of samples that satisfy some error criteria. To find the optimum sampling frequency periodic interpolation for resampling and reconstruction and suitable zero finding of the discrete variable nonlinear equation by combination of the time and frequency computation were used to increase the speed of convergence.

This paper proposes a simplified solution procedure to the model presented by Akinc and Roodman. The Benders’ decomposition procedure for analyzing this model has been developed, and its shortcomings have been highlighted. Here, the special nature of the problem is exploited which allowed us to develop a new algorithm through surrogating method. The two methods are compared by several numerical examples. Computational experience with these data shows the superiority of the new approach. In addition, the required computer programs have been prepared by the authors using TURBO PASCAL 7.0 to execute the algorithm.

A spray casting 7075 alloy with Zn, Mg, Zr,… as dispersoid forming elements has been studied at aging temperature of 120°C. Tensile properties of the alloy at room temperature are higher than that of many high strength aluminum alloys. (Sy = 685 MPa, Su = 698 MPa, after 8hr aging at 120°C). General fatigue behavior (S-N curves) of heat-treated alloy (8hr/120°C) was studied. Endurance limit of the alloy at N = 1.8×106 is 228 Mpa, which is better than that of many traditional aluminum alloys. SEM fractography of fractured fatigue samples showed that the subsurface coarse intermetallic particles are the main source of fatigue crack initiation. Fatigue crack propagation mechanism is dominant transcrystalline mode for low aging temperature and time (8 hr/120°C) but in higher aging temperature regime (3 hr/150°0C) this mechanism changes to intergranular mode.

The method of matched asymptotic expansions, which has been used in previous studies of steady natural convection flow, is extended here to transient natural convection flow at high Prandtl number (Pr). Second-order expansion solutions, valid for large Prandtl numbers, are presented for the transient natural convection flow near a vertical surface which undergoes a step change in temperature. Throughout the transient, the flow is found to have the same dual-layer structure which is characteristic of the steady flow at high Prandtl number. For large Prandtl number, the time to steady state is shown to increase proportional to square root of Pr. The temperature and velocity overshoot, which occurs during the transient at moderate Prandtl number, is shown to disappear as Pr. Uniformly valid expansions for the velocity and temperature profiles near the surface are found to be in good agreement with the numerical solution of the full governing equations for as low as Pr=16. By increase of Prandtl number, the error because of instability in numerical solution of the full governing equations increases and the necessity of using singular perturbation techniques become more obvious.

Two-phase heat transfer is experimentally examined through vertical small diameter tubes, D =1.45 and 2.8 mm using water under a pressure of 50 to 81 kPa and a natural circulation condition. The pool boiling correlation by Stephan-Abdelsalam and the thermosyphon boiling correlation by Imura, et al. predict the measured experimental data in the 2.8 mm tube with an error of -30%. A large heat transfer enhancement in the 1.45 mm tube is caused by the change of heat transfer regime from nucleate boiling to film evaporation. Predictions of heat transfer coefficients by conduction heat transfer through thin liquid film give good results at high heat flux region and the upper elevation of the test section but give lower values at the other elevations. A proposed correlation predicts the measured data of water in annular flow with convective heat transfer regime within an error of? ±25%. The correlation also predicts well the measured heat transfer coefficients of previous experiment with ethanol in the 1.45 mm diameter tube at a pressure of 99 kPa with an error of 25%.

Abstract Two distinct redistribution grids - adaptation techniques, spring analogy and elliptic grid generator are applied to two-dimensional steady, inviscid, shocked flows, and the ability of each technique is examined and compared. Euler equations are solved base on Roe''s Reimann solver approach to simulate supersonic flow around a sphere, transonic flow about an airfoil and supersonic flow in a symmetric diffuser. In redistribution method using spring analogy, the movement of grid points was controlled by forces analogous to tensional and torsional spring forces set between grid points. In elliptic grid generation, the body fitted coordinate were used based on arc length of the grid lines. It is shown that the use of arc length in grid lines instead of the length of straight lines between grids which were used by other, produces better results in one-direction, but introduces some skew ness problem in the generated grids for in both directions. The results show that when expansion or shock waves are only along one direction of the curvilinear coordinate, the use of elliptic equations to produce adapted grid base on the arc length of the grid lines is suitable and produces good results. Specially, it is most suitable when the elliptic equations and flow equations are solved simultaneously. The ability of the adapted grid technique depends upon the flow configurations. For example for flow over a sphere, both techniques provide good results, but for flow over an airfoil, the spring adaptive technique introduces better results. Also, the elliptic grid generator is more suitable for complex flows.