With　the　rapid　development　of　computing　power　and　commercial　computational　fluid　dynamics　(CFD)　packages,　the　accuracy　of　such　simulations　and　their　reliability　are　being　constantly　improved.　This　technology　has　been　widely　used　in　refrigerated　transport　of　apples.　To　improve　the　accuracy　of　the　simulation　results,　a　key　factor　is　to　select　appropriate　CFD　turbulence　models.　This　study　was　based　on　an　experimental　platform　that　was　similar　to　a　refrigerated　vehicle　or　small-scale　cold　storage.　The　objective　of　the　study　was　to　establish　a　comprehensively　verified　3D　CFD　model　of　this　experimental　platform　to　simulate　the　airflow　and　heat　transfer　in　different　unsteady　turbulence　models　and　to　predict　the　temporal-spatial　temperature　and　velocity　variations　during　cooling.　Different　two-equation　eddy　viscosity　turbulence　models,　such　as　standard　κ-ε,　RNG　κ-ε,　Realizable　κ-ε,　standard　κ-ω,　SST　κ-ω,　and　RSM,　were　compared.　Good　agreement　between　model　prediction　and　measured　data　was　obtained.　The　root-mean-square　error　(RMSE)　was　1.049,　1.033,　1.039,　1.037,　1.014,　and　1.064°C　for　standard　κ-ε,　RNG　κ-ε,　Realizable　κ-ε,　standard　κ-ω,　SST　κ-ω,　and　RSM,　respectively.　Therefore,　there　were　no　significant　differences　in　different　turbulence　models　on　simulating　the　temperature　distribution,　and　there　was　similar　accuracy　when　solving　the　energy　equation　on　different　turbulence　models　with　the　two-equation　eddy　viscosity.　The　study　also　revealed　that　the　simulation　values　of　high　Reynolds　numbers　turbulence　model,　e.g.,　standard　κ-ε,　RNG　κ-ε,　Realizable　κ-ε,　and　RSM　were　slightly　lower　than　measured　values　near　the　plane　wall,　and　the　standard　κ-ω　model　predicted　higher　values　than　measured　ones　in　the　area　of　the　fully　developed　turbulence.　After　comparing　the　accuracy　of　six　two-equation　turbulence　models,　the　SST　κ-ω　model　gave　more　accurate　predictions　by　a　comparison　of　the　experimental　and　simulated　results.　This　study　analyzed　the　accuracy　of　predicting　the　3D　spatial　and　temporal　distribution　of　temperature　and　velocity　with　comparison　of　different　two-equation　turbulence　models　inside　the　experimental　platform.　This　research　thus　provided　a　reliable　method　for　better　understanding　and　selecting　CFD　turbulence　models　of　refrigerated　transport　of　apple.　©　2019　International　Society　for　Horticultural　Science.　All　rights　reserved.