Wind Tunnel Analysis of the Airflow through Insect-Proof Screens and Comparison of Their Effect When Installed in a Mediterranean Greenhouse

Abstract

The present work studies the effect of three insect-proof screens with different geometrical and aerodynamic characteristics on the air velocity and temperature inside a Mediterranean multi-span greenhouse with three roof vents and without crops, divided into two independent sectors. First, the insect-proof screens were characterised geometrically by analysing digital images and testing in a low velocity wind tunnel. The wind tunnel tests gave screen discharge coefficient values of Cd,φ of 0.207 for screen 1 (10 × 20 threads·cm(-2); porosity φ = 35.0%), 0.151 for screen 2 (13 × 30 threads·cm(-2); φ = 26.3%) and 0.325 for screen 3 (10 × 20 threads·cm(-2); porosity φ = 36.0%), at an air velocity of 0.25 m·s(-1). Secondly, when screens were installed in the greenhouse, we observed a statistical proportionality between the discharge coefficient at the openings and the air velocity ui measured in the centre of the greenhouse, ui = 0.856 Cd + 0.062 (R² = 0.68 and p-value = 0.012). The inside-outside temperature difference ΔTio diminishes when the inside velocity increases following the statistically significant relationship ΔTio = (-135.85 + 57.88/ui)(0.5) (R² = 0.85 and p-value = 0.0011). Different thread diameters and tension affects the screen thickness, and means that similar porosities may well be associated with very different aerodynamic characteristics. Screens must be characterised by a theoretical function Cd,φ = [(2eμ/Kpρ)·(1/us) + (2eY/Kp(0.5))](-0.5) that relates the discharge coefficient of the screen Cd,φ with the air velocity us. This relationship depends on the three parameters that define the aerodynamic behaviour of porous medium: permeability Kp, inertial factor Y and screen thickness e (and on air temperature that determine its density ρ and viscosity μ). However, for a determined temperature of air, the pressure drop-velocity relationship can be characterised only with two parameters: ΔP = aus² + bus.

Keywords: aerodynamic; greenhouse; insect-proof screen; microclimate; wind tunnel.

Figure 1

Geometric parameters determined using the methodology developed at University of Almería. Digital microscope images of screen 1 (a); screen 2 (b) and screen 3 (c).

Figure 2

Location of the greenhouse sectors and layout of the climate sensors in the greenhouse.

Figure 3

Pressure drops ΔP (Pa) measured in the wind tunnel experiments (a) and pressure drop coefficient due to the insect-proof screen F_φ (b) in function of air velocity through screen u_s (m·s⁻¹) and its reciprocal 1/u_s, respectively. Theoretical (c) and statistical (d) relationships between the discharge coefficient due to the insect-proof screen C_d,φ and u_s. Insect-proof screens 1 (Δ), 2 (□) and 3 (◊); values of C_d,φ for air velocity equal to 0.25, 0.5 and 1.0 m·s⁻¹ (▲, ■, ♦).

Figure 4

Pressure drops ΔP (Pa) in function of air velocity through screen u_s (m·s⁻¹), determined by the wind tunnel experiments for the insect-proof screen 1 (Δ) analyzed in this work, and for two screens with similar aerodynamic behavior but with different aerodynamic characteristics (K_p, Y and e) and different ratio a/b (screens 5 and 9 in Valera et al., 2006 [39] for low air velocities, 0–4 m·s⁻¹).

Figure 5

Temporal evolution of the outside wind speed u_o (―) and inside air velocity u_i (m·s⁻¹) (a) and wind direction ($―$) and normalised velocity u_i/u_o (b). Inside velocities measured in the centre of sector E with screen 1 ($―$), sector W with screen 2 ($―$), sector E with screen 3 ($\text{- - -}$) and sector W with screen 3 ($\text{- - -}$).

Figure 6

Air temperature (°C) (a) and inside-outside air temperature difference ∆T_io (°C) (b) on the days of the experiment. Outside (―); mean values of all the sensors in sector E with screen 1 ($―$), sector W with screen 2 ($―$), sector E with screen 3 ($\text{- - -}$) and sector W with screen 3 ($\text{- - -}$).