Geometric optimization of radiative enclosures using PSO algorithm

Farahmand, Amir-massoud; Payan, S.; Sarvari, S. M. Hosseini

doi:10.1016/j.ijthermalsci.2012.04.024

Cited by 33 publications

(8 citation statements)

References 24 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The inverse problems can be solved by the regularization (explicit) and optimization (implicit) methods . Optimization techniques can be classified as gradient‐based methods and heuristic or gradient‐free methods . In the gradient‐based methods, the local topography of the objective function is used to find a path toward the minimum point of the objective function, that is, by using the first and sometimes the second derivative of the objective function.…”

Section: Introductionmentioning

confidence: 99%

“…For absorbing and emitting medium the RTE is as follows:

(\vec{s} . \nabla) I_{ν} (\vec{r}, \vec{s}) = κ_{ν} (I_{b, ν} (\vec{r}) - I_{ν} (\vec{r}, \vec{s})),

in which

I_{ν} false(\overset{r}{true}, \overset{s}{true} false)

is the spectral radiation intensity at the position

\overset{r}{true}

and in the direction

\overset{s}{true}

and the subscripts

ν

and b denote the wavenumber and blackbody, respectively. In the case of a gray and diffuse surface, the boundary condition for Equation can be expressed as

I_{ν}^{o u t} ({\overset{r}{true}}_{normalw}, \vec{s}) = ε_{normalw} I_{b, ν} ({\overset{r}{true}}_{normalw}) + \frac{(1 - ε_{normalw})}{π} \int_{{\overset{n}{true}}_{normalw} \cdot \vec{s'} < 0} I_{ν, w}^{i n} true| {\overset{n}{true}}_{normalw} \cdot \overset{s}{true} false' true| d normalΩ false',

where

ε_{w}

I_{normalb, ν}

, and

{\vec{n}}_{w}

are ...…”

Section: Introductionmentioning

confidence: 99%

“…11 Optimization techniques can be classified as gradient-based methods 4,8,10,23 and heuristic or gradient-free methods. 19,20,[24][25][26][27] In the gradientbased methods, the local topography of the objective function is used to find a path toward the minimum point of the objective function, that is, by using the first and sometimes the second derivative of the objective function. The heuristic or gradient-free methods use a constructed random sampling of the objective function over the entire feasible region to find the minimum point of the objective function.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Nongray inverse boundary design problems in enclosures at radiative equilibrium

Amiri

Lari

2019

Heat Trans. Asian Res.

View full text Add to dashboard Cite

In this paper, an inverse analysis is used to find an appropriate heat flux distribution over the heater surface of radiant enclosures, filled with nongray media at radiative equilibrium from the knowledge of desired (prespecified) temperature and heat flux distributions over the given design surface. Regular and irregular 2D enclosures filled with nongray combustive gas products are considered. Radiation is considered the dominant mode of heat transfer and the medium temperature is obtained from the energy equation. To evaluate the nongray behavior of the participating gases properly, the spectral‐line weighted‐sum‐of‐gray‐gases (SLW) model with updated correlations is used. The dependence of absorption coefficients and the weights of the SLW model on the temperature of the medium makes the inverse problem nonlinear and difficult to handle. Here, the inverse problem is formulated as an optimization problem and the Levenberg‐Marquardt method has been used to solve it. The finite volume method is exploited for the discretization of the energy equation and the spatial discretization of the radiative transfer equation (RTE). The discrete ordinates method (TN quadrature) is used for the angular discretization of RTE. Five test cases, including homogeneous and inhomogeneous media, are investigated to prove the ability of the present methodology for achieving the desired conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

“…For absorbing and emitting medium the RTE is as follows:

(\vec{s} . \nabla) I_{ν} (\vec{r}, \vec{s}) = κ_{ν} (I_{b, ν} (\vec{r}) - I_{ν} (\vec{r}, \vec{s})),

in which

I_{ν} false(\overset{r}{true}, \overset{s}{true} false)

is the spectral radiation intensity at the position

\overset{r}{true}

and in the direction

\overset{s}{true}

and the subscripts

ν

and b denote the wavenumber and blackbody, respectively. In the case of a gray and diffuse surface, the boundary condition for Equation can be expressed as

I_{ν}^{o u t} ({\overset{r}{true}}_{normalw}, \vec{s}) = ε_{normalw} I_{b, ν} ({\overset{r}{true}}_{normalw}) + \frac{(1 - ε_{normalw})}{π} \int_{{\overset{n}{true}}_{normalw} \cdot \vec{s'} < 0} I_{ν, w}^{i n} true| {\overset{n}{true}}_{normalw} \cdot \overset{s}{true} false' true| d normalΩ false',

where

ε_{w}

I_{normalb, ν}

, and

{\vec{n}}_{w}

are ...…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Nongray inverse boundary design problems in enclosures at radiative equilibrium

Amiri

Lari

2019

Heat Trans. Asian Res.

View full text Add to dashboard Cite

show abstract

“…From previous studies, they have been implemented in mostly inverse gas radiation analyses concerned with the determination of the unknown radiation properties of a gas medium, boundary conditions and temperature profiles from given radiation measurements [6][7][8][9][10][11]. For surface radiation problems containing transparent media, there are some works concerning the application of GA and PSO algorithms to the optimal geometry design of radiant enclosures inversely [12][13][14][15][16]. Hosseini Sarvari [12] applied the micro-genetic algorithm (mGA) to optimize the geometry of an enclosure which contains a transparent medium, and Safavinejad et al [13,14] used mGA for determining the optimal number and location of heaters over boundary surfaces in irregular 2-D transparent media.…”

Section: Introductionmentioning

confidence: 99%

“…Hosseini Sarvari [12] applied the micro-genetic algorithm (mGA) to optimize the geometry of an enclosure which contains a transparent medium, and Safavinejad et al [13,14] used mGA for determining the optimal number and location of heaters over boundary surfaces in irregular 2-D transparent media. Also, Chopade et al [15] estimated heat flux distributions on a 3-D design object using mGA, while Farahmand et al [16] studied the shape optimization of a two-dimensional radiative enclosure with diffuse-gray surfaces by the PSO algorithm and compared with mGA results from Sarvari [12] briefly. However, to the author's best knowledge, up to now it is surveyed that GA and PSO have rarely been implemented in the inverse estimation studies of unknown surface radiative properties when no participating media is involved, except Kim and Baek [17] used hybrid GA for retrieving surface emissivity and its temperature in an axisymmetric cylinder.…”

Section: Introductionmentioning

confidence: 99%