Maroun Nemer

Topology Optimization of Heat and Mass Transfer Problems: Laminar Flow

The design of efficient structures for heat and mass transfer problems involves the implementation of an appropriate topology optimization strategy in order to fully take into account the bi-objective nature of the problem. This article couples the finite-volume method (FVM), for the direct solver, with the discrete adjoint approach, for the sensitivity analysis, in order to tackle both fluid dynamic and heat transfer optimization in the frame of laminar flows. Details are provided about the sparsity pattern of the discrete adjoint system, which requires special attention to select a suitable matrix iterative solver. Several examples underline the adequacy of topology optimization in conjunction with the FVM for the minimization of the power dissipated by the fluid. Then, a bi-objective problem aiming at minimizing the pressure drop while maximizing the recoverable thermal power is solved by the identification of its Pareto frontier, thanks to an aggregate objective function (AOF) method. The main conclusion deals with the possibility of finding an acceptable trade-off between both objectives and the potential of topology optimization for heat and mass transfer optimization.

Topology Optimization Using the SIMP Method for Multiobjective Conductive Problems

The efficient cooling of a finite-size volume generating heat, including adiabatic boundary conditions with the exception of a small heat sink, poses the problem of optimal allocation of high-conductivity material. Among the structural optimization methods, this article couples solid isotropic material with penalization parametrization (SIMP) with an aggregated objective function approach (AOF) to tackle this topology optimization problem through a multiobjective strategy. Both average and variance temperature-reduction problems is solved by the identification of Pareto fronts, which are highly dependent on the quantity of the high-conductivity material. This study also underlines the link between the sensitivity analysis of both objective functions, which is required by the method of moving asymptotes (MMA). Furthermore, additional calculations have been done to include variations in heat-generation rate between two conductive materials by means of an additional penalization strategy. The main conclusion deals with the possibility of finding an acceptable trade-off between average and variance objective functions thanks to the convex shape of Pareto frontiers.

Transient Radiation and Conduction Heat Transfer in Glass Sheets by the Thin Layer Approximation

This paper is devoted to the simulation of 3D transient radiation and conduction heat transfer occurring inside thin glass sheets undergoing high temperature processing. The glass is considered as an absorbing, emitting, and nonscattering medium. The zonal method is used to establish the governing radiation transfer model. Direct exchange areas are calculated by the flux planes approximation. The thin layer approximation (TLA) is then introduced for increasing CPU efficiency. Three different numerical integration schemes made possible by the TLA are presented. Comparisons are made, with calculations performed using the finite volume method (FVM). The transient coupled energy equation is solved by a full implicit control volume method using the incomplete Cholesky conjugate gradient method. The heat transfer analysis of a glass sheet residing inside a hot rectangular enclosure is studied. Results obtained by the zonal method, with or without the TLA, are in close agreement with those obtained by the FVM. CPU requirements for radiative heat transfer analysis of the zonal method with TLA are, depending on the numerical integration scheme used, between 8 and 23 times smaller than those of the zonal method without TLA. The difference between the results of the different models never exceeds 4%. The zonal method with the TLA offered significant improvements in CPU time when compared with the original zonal method with similar or acceptable accuracy.

The Re-Plating Algorithm for Radiation Total Exchange Area Calculation

Great efforts have been made to date toward modeling nongray radiative heat transfer accurately. In this article, a new version of the plating algorithm, designated the re-plating algorithm, for total exchange areas (TEAs) calculation from direct exchange areas (DEAs) for nongray radiative problems is presented. The re-plating algorithm calculates TEAs for a given band number b from those of band number b − 1 by performing successive re-plating procedures. The effectiveness of the new algorithm is demonstrated for thermal modeling of an aluminum brazing furnace and a glass treatment furnace. CPU requirements for TEA calculation were reduced significantly.

Experimental Validation of the Multiple Absorption Coefficient Zonal Method (MACZM) in a Dynamic Modeling of a Steel Reheating Furnace

In this work, the multiple absorption coefficient zonal method (MACZM) is being implemented and validated numerically. The method is demonstrated to be highly suitable for the analysis of radiative heat transfer in multidimensional inhomogeneous non-grey media. A uniform rectangular fine grid is considered and small CPU time is achieved. This makes the method of great interest for transient applications. The validity of the method is demonstrated in two steps. First, cases with simple geometry are considered and results are compared to results generated by direct numerical integration. Results are also generated by MODRAY, which is a source project based on an original method called the flux-planes approximation, and are shown to be equally accurate. Second, the case of a steel reheating furnace is considered. In a previous work, the furnace heat balance and temperature profiles were simulated using a finite difference computation approach and radiative exchange factors generated by MODRAY. Experiments were performed and results generated by the model were found to be in good agreement with experimental data. The radiative exchange factors are now recalculated with MACZM. They are shown to be very close to those generated by MODRAY. The comparison of the two methods clearly shows that MACZM is much faster for the calculation of the volume-volume radiative exchange factors on a uniform rectangular grid.

Optimization of radiant heater power for heating of flat plates using metaheuristic methods

In this article, simulated annealing (SA), genetic algorithm (GA), and tabu search (TS) methods are employed separately and together for finding optimal radiant heater settings for obtaining a specific temperature profile on a design surface. Results show that all methods yield to very acceptable results for various desired temperature profiles but after different number of iterations. The GA and TS methods progress very rapidly before slowly stabilizing after a certain number of iterations, while the SA method progresses for longer periods. In addition, hybrid combinations of GA, SA, and TS proved to be better than separate ones.

The component interaction network approach for modeling of complex thermal systems

A practical approach for the thermal modeling of complex thermal systems, called the component interaction network (CIN) is presented. Its stages are explained: description of the thermal system as a set of non-overlapping components and their interactions by heat and mass exchanges, modeling of components with different levels of precision using finite volumes and finite elements, modeling of interactions by conduction, convection, radiation and advection, time resolution scheme and simulation. Non-conventional notions of conditional existence of components or time events are introduced. The approach is illustrated with a simple example of an electric furnace. It is then applied to a rapid thermal processing (RTP) furnace and validated experimentally. The advantages of the CIN approach are demonstrated.

An Efficient CPU-GPU Implementation of the Multiple Absorption Coefficient Zonal Method (MACZM)

The multiple absorption coefficient zonal method (MACZM) is an efficient radiative heat transfer modeling method in nonisothermal inhomogeneous media. The method is of high interest for dynamic applications because of its ability to asses semitransparent radiative heat transfer in very short computation time. In this work, an efficient algorithm for MACZM is implemented. A connectivity control study is presented for taking into account the connectivity considerations required by the method. An identified ray traversal algorithm corresponding to part of the MACZM implementation is then selected among three different approaches presented in the article, based on well-known ray traver sal algorithms, the 6-tripod line algorithm and the 6-parametric line algorithm. On the other hand, the MACZM is highly parallel and is implemented in CUDA, a parallel computing architecture that enables easy use of a powerful graphics processing unit (GPU). An efficient implementation is discussed consisting of an optimal solution for exploiting the method parallelism (threading) and the use of the memory resources available on the GPU. Speed-ups going from 300 to 600 times are achieved, using a NVIDIA Tesla C 1060 GPU and an Intel Xeon CPU E5507 at 2.27 GHz. Radiative heat transfer is then simulated in a steel reheating furnace using the optimized GPU implementation. The computation time is further reduced by using a multigrid approach.

Modified Zonal Method for Thin Solid Semi-Transparent Media With Reflective Boundary

In this paper a detailed description of a modified zonal method used for the prediction of transient operation in glass treatment furnaces is presented. This method calculates the radiative transfer factors for scenes containing both diffusive surfaces and absorbing-emitting/non-scattering media with high indices of refraction all embedded in a totally transparent atmosphere. The method used combines the flux plans approximation for calculation of the view factors, the plating algorithm originally developed by Edwards for calculation of the total heat transfer factors, and finally, Emery’s equations for deduction of the surface-volume and volume-volume heat transfer factors. The matrix of radiative transfer is then simplified by eliminating the low energy level factors, thus rendering the matrix hollow and thereafter decreasing the time required for calculation.Copyright

Fuel Consumption Saving Potential of Stirling Machine on Series Parallel Hybrid Electric Vehicle: Case of the Toyota Prius

Investigations on alternative fuels and new hybrid powertrain architectures have recently undergone significant e fforts in the automotive industry, in attempt to reduce carbon em issions from passenger cars. The use of these fuels presents a p o ential for reemerging the deployment of external combustion nonc ventional engines in automotive applications, such as the Sti rling engines, especially under the current development context of powertrain electrification. This paper investigates the potent ial of fuel consumption savings of a series-parallel hybrid ele ctric vehicle (SPHEV) using a Stirling machine as fuel converter. An exergotechnological explicit analysis is conducted to ide ntify the Stirling system configuration presenting the best compromise between high efficiency and automotive implementation constraint s. The Stirling engine with combustion chamber preheater is priorit ized. A SPHEV model is developed based on the Prius power-split h ybrid electric architecture. Energy consumption simulations are pe formed on the worldwide-harmonized light vehicles test cycle (WLT C) using dynamic programing as global optimal energy managem ent strategy. Results show improved fuel consumption performance of the Stirling machine compared to the ICE. In addition, the Stirl ing offers other intrinsic advantages such as low noise and vibratio n operation and mainly multi-fuel use capability. Consequently, the studied Stirling presents a potential for implementation on SPHEVs