SIMULATION OF CONJUGATE CONVECTIVE-CONDUCTIVE HEAT TRANSFER IN A MICROCHANNEL WITHIN THE SLIP REGIME USING GPU ACCELERATED LATTICE BOLTZMANN METHOD

Widyaparaga, Adhika and Pranowo, . SIMULATION OF CONJUGATE CONVECTIVE-CONDUCTIVE HEAT TRANSFER IN A MICROCHANNEL WITHIN THE SLIP REGIME USING GPU ACCELERATED LATTICE BOLTZMANN METHOD. In: ASME 2013 11th International Conference on Nanochannels, Microchannels, and Minichannels ICNMM2013, June 16-19, 2013, Sapporo, Japan.

	Text (Prosiding_Internasional terindeks) C5_07_ICNMM_2013.pdf Download (1MB)
	Text Peer_Review_C5_07_ICNMM_2013.pdf Download (551kB)
	Text Cek_Turnitin_C5_07_ICNMM_2013.pdf Download (3MB)

Abstract

This study aims to investigate the significance of conjugate heat transfer in the microscale within the slip regime. As within the slip regime the continuum assumption is invalid due to presence of rarefaction effects, the Lattice Boltzmann method (LBM) is employed to overcome the limitations of Navier Stokes based solutions in this regime. We have constructed and compared two case models in which a fluid of higher temperature enters a microchannel. The conditions are set to obtain Knudsen numbers which result in the slip regime being dominant. To investigate the effect of conjugate heat transfer, the two models differed in the aspect that one model did not incorporate conjugate heat transfer and while the other did. The numerical calculation was validated by comparing the velocity profile results to exact theoretical approximations and was found to agree well. The results of comparison of models Case I and Case 2 have shown that temperature profile is affected significantly by conjugate heat transfer. The conjugate heat transfer at the microchannel wall (Case 1) was shown to maintain the initial temperature of fluid longer than compared to a purely isothermal wall (Case 2), thus signifying the importance of the consideration of conjugate heat transfer effects in microfluid models. We have implemented GPU based parallel processing to reduce computation time. The result of the incorporation of GPU processing was found to increase processing speed up to 15 times.

Actions (login required)

View Item