A Bi-Layer compact thermal model for uniform chip temperature control with non-uniform heat sources by genetic-algorithm optimized microchannel cooling

Wu, Ruikang; Zhang, Xinfeng; Fan, Yiwen; Hu, Run; Luo, Xiaobing

doi:10.1016/j.ijthermalsci.2018.10.047

Cited by 26 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their results showed that the thermal resistance was effectively reduced by 0.144 W/K. Wu et al [106] proposed a two-layer compact microchannel model with a non-uniform heat source, they used the GA prediction model to obtain the best parameters for fluid velocity and fluid temperature and set the channel width as the prediction parameter. Their outputs revealed that the model can accurately predict the microchannel temperature, and the optimized model has uniform surface temperature distribution under extremely uneven heat sources.…”

Section: Taguchi Methods (Tm) and Genetic Algorithms (Ga)mentioning

confidence: 99%

Fluid flow and heat transfer in microchannel heat sinks: Modelling review and recent progress

Gao

Yang

et al. 2022

Thermal Science and Engineering Progress

View full text Add to dashboard Cite

Section: Taguchi Methods (Tm) and Genetic Algorithms (Ga)mentioning

confidence: 99%

Fluid flow and heat transfer in microchannel heat sinks: Modelling review and recent progress

Gao

Yang

et al. 2022

Thermal Science and Engineering Progress

View full text Add to dashboard Cite

“…Multi-objective genetic algorithm (MOGA) was used to investigate the thermal and hydrodynamic performance of a BNN nanofluid-cooled MCHS at various concentrations of the BNN at 50C. MOGA has been frequently used in the performance modelling of a MCHS [25][26][27][28]. Furthermore, thermophysical properties issued from experiments have been used for providing reliability to the optimization modelling outcomes which can lead to the best concentration identification for the MCHS [29].…”

Section: Introductionmentioning

confidence: 99%

Optimization of a boron nitride nanotubes nanofluid-cooled microchannel heat sink at different concentrations

Yusof

Shamsuddin

Estellé

et al. 2022

J Therm Anal Calorim

View full text Add to dashboard Cite

The microchannel heat sink (MCHS) has become the most relevant micro-heat exchanger for a small area in need of an effective high heat removal system. However, heat dissipation from the microchips where the MCHS is utilized -the microprocessor and microcontroller -are getting higher with the sizes getting smaller. A more effective coolant is needed to address the increasing heat load from the microchip and nanofluid, nanosized particles dispersed in a base fluid, is among those explored. This paper reports a new potential use of non-metallic nanofluid, boron nitride nanotube (BNN) that is capable of improving the overall performance of a rectangular MCHS. A heuristic method, multi-objective genetic algorithm (MOGA), is employed to simultaneously minimize the thermal resistance and pressured drop of a boron nitride nanotubes (BNN) nanofluid-cooled MCHS to obtain optimized dimensions at various weight concentrations. The method is capable in achieving conflicting objectives, minimization of the thermal resistance and the pressure drop; an increase in the former decreases the latter and vice versa. In addition, experimental thermophysical properties of the BNN nanofluid are used to provide reliability to the optimization outcomes in identifying the best BNN concentration for cooling of a MCHS at 50°C. The optimization results showed that as the thermal resistance decreases, the pressure drop decreases.

show abstract

“…16,17 These optimizations differ in terms of the design variables. For cooling channels, the design variables of sizing optimization can be considered as the cross-sectional dimensions of the channels, 18 and the design variables of shape optimization can be considered as the control parameters of the structure boundary curve or surface shape, such as the node coordinates of the channels. Tan et al 19 used the channel control points as the design parameters, which define the network shape.…”

Section: Introductionmentioning

confidence: 99%

“…In these previous studies, different topology optimization methods were used in the optimization of cooling channels, such as the density-based method, 25,27,29,30 LSM, 26,28 and moving morphable component-based approach. 31,32 Moreover, different numerical optimization algorithms have been adopted to update design variables, such as sequential linear programming (SLP), 25 sequential quadratic programming (SQP), 29,30 method of moving asymptote (MMA), 27,31 and GA. 18,24,33 In general, numerical optimization algorithms can be classified into two categories: one involves the use of derivative information to construct search algorithms, such as SLP and MMA 34 ; the other involves the construction of a search algorithm solely based on the calculation of the objective function, such as heuristic random search algorithms. Statistical laws based on random search are typically introduced to improve the algorithm efficiency of heuristic algorithms.…”

Section: Introductionmentioning

confidence: 99%

Net‐based thermal‐fluid model and hybrid optimization of cooling channels

Liu

Wang

et al. 2022

Numerical Meth Engineering

View full text Add to dashboard Cite

This study aims to propose a hybrid optimization method for cooling channels based on the improved net-based thermal-fluid model. The channel topology was optimized through the genetic algorithm (GA), whereas the channel diameter and node position were optimized using the moving asymptote method (MMA). The linear combination of the average temperature and pressure drop was defined as the objective function. Three-dimensional computational fluid dynamics (CFD) simulations were performed to verify the physical fields of the optimized results. The optimized results indicated that, as the weighting factor (w) for the average temperature increased, the flow distribution was more reasonable and uniform, effectively decreasing the average temperature at the cost of higher pressure drops. In comparison with the optimization entirely based on GA, the hybrid optimization could yield a similar and efficient cooling performance with lower time-consuming. Moreover, the optimized results obtained under the suitable objective could achieve the better thermal performance at a lower pressure drop in comparison with the conventional serpentine-shaped channel layout. Finally, the hybrid optimization method used in this article was compared with the conventional density approach, and the former could obtain the clear and complex channel distribution with greatly reduced computational time.

show abstract

A Bi-Layer compact thermal model for uniform chip temperature control with non-uniform heat sources by genetic-algorithm optimized microchannel cooling

Cited by 26 publications

References 30 publications

Fluid flow and heat transfer in microchannel heat sinks: Modelling review and recent progress

Fluid flow and heat transfer in microchannel heat sinks: Modelling review and recent progress

Optimization of a boron nitride nanotubes nanofluid-cooled microchannel heat sink at different concentrations

Net‐based thermal‐fluid model and hybrid optimization of cooling channels

Contact Info

Product

Resources

About