Analysis of Microclimate Uniformity in a Naturally Vented Greenhouse with a High-Pressure Fogging System

Abstract

<abstract><title><italic>Abstract.</italic></title> Controlled environments have long played an important role in the field of agriculture because they enable growers to more precisely regulate not only the quantity of a crop but the quality as well. Currently, most controlled environments rely on mechanical systems to regulate temperature and relative humidity within the controlled space, but these systems can be costly to install and operate. For this reason, researchers have begun investigating the efficacy of using high-pressure fogging systems and their associated control strategies in naturally ventilated greenhouses as an alternative to mechanical cooling methods. However, conducting detailed analyses of a greenhouse’s aerodynamics requires carefully arrayed instruments, a consideration of many different scenarios, and a significant amount of time to compile data, not to mention the monetary cost of experimental analysis. The objective of this study, then, was to develop a 3D computational fluid dynamics (CFD) model capable of more efficiently analyzing the movement of air in a naturally ventilated greenhouse equipped with a high-pressure fogging system. The overall model included five subunits: (1) a porous media model to simulate the ways that a crop canopy will affect airflow, (2) a solar load model and (3) a discrete ordinates radiation model to simulate solar radiation, (4) a species transport and discrete phase model to simulate evaporation of droplets, and (5) an evapotranspiration (ET) model integrated with a user-defined function (UDF). The overall model predicted temperature and relative humidity within the greenhouse with percentage errors for temperature and relative humidity of 5.7% to 9.4% and 12.2% to 26.9%, respectively (given a 95% confidence interval). The average percent error between the simulated and measured ET was around 8%, and the CFD-simulated stomatal and aerodynamic resistances agreed well and were within the ranges indicated by earlier research. Having validated the overall model with experimental data, we then used a 2⁴ full-factorial design to determine the effects on climate uniformity produced by four factors: position of the side vent, position of the vertical sprayer nozzles, position of the horizontal sprayer nozzles, and angle of the nozzle. On the basis of our statistical analysis, we concluded that “horizontal nozzle position” was the most significant factor for climate uniformity, while the least significant factor among those evaluated was “side vent opening.”