MATHEMATICAL OPTIMIZATION: APPLICATION TO THE DESIGN OF OPTIMAL MICRO-CHANNEL HEAT SINKS
Abstract
This paper documents the geometrical optimization of a micro-channel heat
sink embedded inside a highly conductive solid, with the intent of developing optimal solutions for thermal management in microelectronic devices. The objective is to minimize the peak wall temperature of the heat sink subject to various constraints such as manufacturing restraints, fixed pressure drop and total fixed volume. A gradient based multi-variable optimization algorithm is used as it adequately handles the numerical objective function obtained from the computational fluid dynamics simulation. Optimal geometric parameters defining the micro-channel were obtained for a pressure drop ranging from 10 kPa to 60 kPa corresponding to a dimensionless pressure drop of 6.5 × 107 to 4 × 108 for fixed volumes ranging from 0.7 mm3 of 0.9 mm3. The effect of pressure drop on the aspect ratio, solid volume fraction, channel hydraulic diameter and the minimized peak temperature are reported. Results also show that as the dimensionless pressure drop increases the maximised dimensionless global thermal conductance also increases. These results are in agreement with previous work found in literature.
sink embedded inside a highly conductive solid, with the intent of developing optimal solutions for thermal management in microelectronic devices. The objective is to minimize the peak wall temperature of the heat sink subject to various constraints such as manufacturing restraints, fixed pressure drop and total fixed volume. A gradient based multi-variable optimization algorithm is used as it adequately handles the numerical objective function obtained from the computational fluid dynamics simulation. Optimal geometric parameters defining the micro-channel were obtained for a pressure drop ranging from 10 kPa to 60 kPa corresponding to a dimensionless pressure drop of 6.5 × 107 to 4 × 108 for fixed volumes ranging from 0.7 mm3 of 0.9 mm3. The effect of pressure drop on the aspect ratio, solid volume fraction, channel hydraulic diameter and the minimized peak temperature are reported. Results also show that as the dimensionless pressure drop increases the maximised dimensionless global thermal conductance also increases. These results are in agreement with previous work found in literature.
Keywords
micro-channel; geometric optimization; thermal conductance; gradient based algorithm
Full Text:
PDFDOI: http://dx.doi.org/10.5380/reterm.v8i1.61883
