Numerical thermal analysis and optimization of multi-chip LED module using response surface methodology and genetic algorithm

Hong-Yu Tang, Huai-Yu Ye, Xian-Ping Chen, Cheng Qian, Xue-Jun Fan, Guo-Qi Zhang*

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

13 Citations (Scopus)
83 Downloads (Pure)

Abstract

In this paper, the heat transfer performance of the multi-chip (MC) LED module is investigated numerically by using a general analytical solution. The configuration of the module is optimized with genetic algorithm (GA) combined with a response surface methodology. The space between chips, the thickness of the metal core printed circuit board (MCPCB), and the thickness of the base plate are considered as three optimal parameters, while the total thermal resistance (Rtot) is considered as a single objective function. After optimizing objectives with GA, the optimal design parameters of three types of MC LED modules are determined. The results show that the thickness of MCPCB has a stronger influence on the total thermal resistance than other parameters. In addition, the sensitivity analysis is performed based on the optimum data. It reveals thatRtot increases with the increased thickness of MCPCB, and reduces as the space between chips increases. The effect of the thickness of base plate is far less than that of the thickness of MCPCB. After optimization, three types of MC LED modules obtain lower Tj andRtot. Moreover, the optimized modules can emit large luminous energy under high-power input conditions. Therefore, the optimization results are of great significance in the selection of configuration parameters to improve the performance of the MC LED module.

Original languageEnglish
Article number8006225
Pages (from-to)16459-16468
Number of pages10
JournalIEEE Access
Volume5
DOIs
Publication statusPublished - 9 Aug 2017

Keywords

  • genetic algorithm
  • Multi-chip LED module
  • optimization
  • response surface methodology
  • thermal resistance
  • OA-Fund TU Delft

Fingerprint

Dive into the research topics of 'Numerical thermal analysis and optimization of multi-chip LED module using response surface methodology and genetic algorithm'. Together they form a unique fingerprint.

Cite this