摘要:

A numerical analysis is performed on thermocapillary buoyancy convection induced by phase change in a liquid droplet. A finite-difference code is developed using an alternating-direction implicit (ADI) scheme. The intercoupling relation between thermocapillary force, buoyancy force, fluid property, heat transfer, and phase change, along with their effects on the induced flow patterns, are disclosed. The flow is classified into three types: thermocapillary, buoyancy, and combined convection. Among the three mechanisms, the combined convection simulates the experimental observations quite well, and the basic mechanism of the observed convection inside evaporating sessile drops is thus identified. It is disclosed that evaporation initiates unstable convection, while condensation always brings about a stable density distribution which eventually damps out all fluid disturbances. Another numerical model is presented to study the effect of boundary recession due to evaporation, and the “peeling-off” effect (the removal of the surface layer of fluid by evaporation) is shown to be relevant.

ISSN:1040-7782    

DOI:10.1080/10407788908944710

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor

摘要:

This study investigates the heat transfer mechanisms of a uniform wall heat flux wedge located in an air-water mist flow. In order to yield a physical model closer to the realistic situation, two important factors are considered, namely droplet trajectories and the size distribution of polydisperse droplets. An equivalent diameter is adopted to economize computing time. An effective numerical method is proposed to solve the governing equations simultaneously. The effect of the three kinds of heat transfer mechanisms—gas-phase convective heat transfer, water film sensible heat transfer, and evaporation heat transfer—on the heat transfer coefficient are examined in detail.

ISSN:1040-7782    

DOI:10.1080/10407788908944711

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor

摘要:

Thermocapilktry migration of a spherical bubble under microgravity in an infinite medium with a linear temperature distribution is investigated. The two-dimensional Navier-Stokes equations for axisymmetric flow around the bubble are discretized in polar coordinates on a ring around the bubble. The grid has equally spaced mesh lengths Δr in the r direction and Δθ in the θ direction. Velocities and temperature are calculated at the vertices of the quadrilateral cells of this grid. Pressure is calculated at the cell centers. The energy equation and the two momentum equations are discretized by standard three-point second-order central differences centered at the cell vertices. The continuity equation is discretized by two-point second-order central differences around the centers of the cells. The resulting nonlinear system of algebraic equations for velocities, temperature, pressure, and the terminal steady-state velocity of the bubble are solved by an iterative procedure that uses the Newton and secant methods. The banded linear systems that occur are solved by a direct solver. Computations have been carried out for Reynolds numbers from 10−7to 2000 and Marangoni numbers from 10−7to 1000. For these ranges, the scaled bubble velocities vary by less than one order of magnitude. The bubble velocity is influenced more by the Marangoni number than by the Reynolds number.

ISSN:1040-7782    

DOI:10.1080/10407788908944712

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor

摘要:

The turbulent flow about a two-dimensional blunt rectangular section is used as a test case to examine the performance of seven near-wall turbulence models. The first two models are one-equation low Reynolds number (Re) models requiring a fine grid near-wall treatment. The other near-wall turbulence models considered are based on wall functions that bridge with a single cell the thermally important near-wall region. Standard wall functions based on the local equilibrium assumption, wall functions using a two- and three-layer approach to evaluate local variations of turbulence quantities in the k equation, and extension of the two- and three-layer treatments to the ε equation are considered. The numerical predictions are obtained using a variant of the k-ε model incorporating a curvature correction. The governing equations are discretized using a finite-volume formulation employing the bounded-skew hybrid differencing scheme. The solutions are obtained using a two-pass procedure, devised to allow for the correct use of the wall functions. Computations are performed for Re in the range 20,000-75,000. The various near-watt models are assessed by comparing resulting Nusselt number distributions and selected fluid dynamic results with available experimental data. A three-layer model of the wall region, applied to the k and ε equations, gives good agreement with the data; the standard wall Junction treatment is not satisfactory at all for this flow.

ISSN:1040-7782    

DOI:10.1080/10407788908944713

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor

摘要:

A theoretical investigation of one-dimensional forced convection heat transfer in He II is conducted. The problem of interest involves a flow tube containing He II, which is heated at its midpoint along its length. Two modes of heating are analyzed: step function and square pulse. The one-dimensional He II energy equation is used to find the temperature distribution along the tube for both steady-state and transient situations. For the steady-state case, a numerical integration routine is used to obtain a solution, whereas for the solution of the transient case, a finite-difference scheme is developed. The numerical temperature profiles are then shown to compare well with the results of an experiment.

ISSN:1040-7782    

DOI:10.1080/10407788908944714

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor

摘要:

This paper describes an iterative numerical optimization procedure for generating the cryosurgical probe tip geometry to produce the desired lethal temperature envelope for steady-state axisymmetric systems. The Kirchhoff transformation is used to include the nonlinear effect of variable thermal conductivity at cryogenic temperatures. The boundary-element method (BEM) is used to solve the governing differential equation at each step in the iteration. The shape optimization procedure involves sequential searches of radial vectors to determine the optimum location for each node at each step in the iteration. Experimental data are obtained for three probe tip geometries, and the experimental data compare favorably with the results of the numerical solution.

ISSN:1040-7782    

DOI:10.1080/10407788908944715

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor

摘要:

An inverse heat conduction method for simultaneously estimating spatially varying thermal conductivity and heat capacity per unit volume under the conditions of a flash method type of experiment is developed. The unknown thermal properties are assumed to vary only in the space dimension normal to the slab sample and are modeled with piecewise linear representations. Lacking in the literature are specific requirements that must be satisfied by the number of measurements in the spatial domain in order to ensure uniqueness of the inverse solution. We prepared a series of numerical experiments to provide a better understanding of this issue. Multiple temperature sensors are shown to be necessary to determine spatially varying properties. The effectiveness of the method is illustrated through simulated experimental applications of the method.

ISSN:1040-7782    

DOI:10.1080/10407788908944716

出版商:Taylor & Francis Group    

年代:1989

数据来源: Taylor