agronomy
Article
CFD Analysis and Optimization of a Plastic Greenhouse with
a Semi-Open Roof in a Tropical Area
Haoran Yin 1 , Kaiji Wang 1 , Jiadong Zeng 2 and Zhenzhen Pang 1, *
Citation: Yin, H.; Wang, K.; Zeng, J.;
Pang, Z. CFD Analysis and
Optimization of a Plastic Greenhouse
with a Semi-Open Roof in a Tropical
Area. Agronomy 2024, 14, 876. https://
doi.org/10.3390/agronomy14040876
Academic Editor: Jorge Antonio
Sánchez-Molina
Received: 5 March 2024
Revised: 29 March 2024
Accepted: 5 April 2024
Published: 22 April 2024
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Agronomy 2024, 14, 876. https://doi.org/10.3390/agronomy14040876
1 School of Breeding and Multiplication (Sanya Institute of Breeding and Multiplication), Hainan University,
Sanya 572022, China; 22210902000011@hainanu.edu.cn (H.Y.); kaijiwang163@163.com (K.W.)
2 School of Civil and Architectural Engineering, Hainan University, Haikou 570228, China;
zjd_rcwe@hainu.edu.cn
* Correspondence: luckypzz@163.com
Abstract: A numerical simulation model of a natural ventilation greenhouse is helpful for improving
the production and quality of greenhouse crops in tropical areas. Field experiments show that the
mean coefficient of variation of indoor light intensity in four seasons was lower than 10.0%. The
highest indoor temperature reached 39.3 ◦ C during summer, while the average indoor temperature
ranged from 24 ◦ C to 26 ◦ C in the other three seasons. The average relative humidity in the greenhouse
ranged from 76% to 87% annually, which was higher and more stable than that in the external
environment. A three-dimensional steady-state numerical model of the greenhouse was established
based on computational fluid dynamics. Under natural ventilation conditions, the maximum error
between the simulated value and the measured value of the temperature in each measuring point was
5.90%. And the average relative error between the simulated and measured values was 3.0% in the
range of 0.7−1.5 m of crop cultivation height. Finally, a numerical simulation of adding side windows
and expanding the vents was carried out. The results show that these methods can homogenize the
airflow distribution in the greenhouse and improve the utilization efficiency of natural ventilation
without more mechanical system operations.
Keywords: CFD simulation; tropical climate; greenhouse; semi-open roof; natural ventilation
1. Introduction
Hainan’s tropical climate, with high temperatures and humidity and heavy storms
in the summer, greatly limits open-field cultivation, so it is of great significance to de-
velop tropical greenhouses in this area. Greenhouse plant quality and yield are affected
by environmental factors [1], and the indoor environment of greenhouses is unevenly
distributed [2]. In order to ensure the intensive production of plants, it is necessary to
regulate the greenhouse environment to meet the growth needs of plants [3]. In tropical
areas, ventilation in greenhouses is the main solution for environmental control. Compared
with mechanical ventilation, natural ventilation is a simple and low-cost method [4] and
requires a lower level of energy consumption. However, the natural ventilation in tropical
greenhouses needs further analysis because of its complexity and uncertainty [5], which
increase the technical difficulty of field experiments and simulation.
The microclimate in natural ventilation greenhouses is highly dependent on the local
climate and the main characteristics of the greenhouse [6]. The efficiency of a natural
ventilation greenhouse is affected by the shape of the roof, the location and size of the
vents, and the external wind speed and direction. Soussi et al. [7] summarized the impact
of natural ventilation on the microclimate in a tropical greenhouse.
Tawalbeh et al. [8] reviewed the research methods and techniques for protected agri-
culture microclimates in tropical and subtropical areas, among which computational fluid
dynamics (CFD) simulation is widely used, providing better accuracy and cost performance
compared with traditional methods. Wu et al. [9] studied a Chinese solar greenhouse (CSG)
https://www.mdpi.com/journal/agronomy
Agronomy 2024, 14, 876 2 of 21

and established a mathematical model of the three-dimensional dynamic thermal environ-

ment at the crop layer by combining experiments with numerical simulations. Compared
with the measured results, the difference between the simulated average temperature and
the measured results was from 0.1 ◦ C to 2.2 ◦ C, and the maximum error was less than
13.20%. The model accurately describes the CSG heat transfer process. Liu et al. [10]
optimized the traditional multi-arch greenhouse structure, designed a crossed multi-arch
greenhouse, and carried out a ventilation simulation analysis. The ventilation and tem-
perature distribution could indicate the natural ventilation thermal condition and provide
a model basis for structural optimization. McCartney et al. [11] adopted a one-fourth
greenhouse model and CFD simulation to analyze the temperature distribution and airflow
velocity in a natural ventilation augmented cooling (NVAC) greenhouse. However, the
boundary effect and multivariable complexity will cause higher system error when the
simulated results are applied to a real greenhouse. In addition, Flores-Velázquez et al. [12]
assessed the thermal load of tomato greenhouses based on Greenhouse Thermal Effective-
ness (GTE) indicators of regional thermal potential, which proposed a concise numerical
model to assess the temperature control needs of greenhouses. Those results inspired the
optimization method for natural ventilation greenhouses in tropical areas.
Therefore, this study determined the variation in air temperature, relative humidity,
and light intensity in a plastic greenhouse with a semi-open roof in Hainan through field
experiments; then, based on the real greenhouse environmental data, a CFD model was
built, and the greenhouse structure under natural ventilation was optimized.

2. Materials and Methods

2.1. Experimental Material
As shown in Figure 1, the experiment was carried out in a plastic greenhouse with a
semi-open roof at Lingshui Modern Agriculture Demonstration Base (109◦ 50′ E, 18◦ 26′ N)
from 2016 to 2017. The greenhouse was oriented north to south, with size dimensions of
145 m (L) × 432 m (W) × 7.2 m (H), and the experimental area was in the middle of the
greenhouse with 9 spans. The covering material was age-resistant polyethylene film with
0.15 mm thickness. The greenhouse roof was covered with a black shading net, with a height
of 8.5 m. Electric film-rolling side windows were set around the greenhouse. Adjustable
Agronomy 2024, 14, x FOR PEER REVIEW 3 of 22
vent windows were installed on the roof, and the window openings were covered by
insect-proof nets. Additionally, circulating fans were set up inside the greenhouse.

Figure1.
Figure 1. Experimental
Experimental greenhouse.
greenhouse.

Natural ventilation
The layout combined
of the indoor with themeasuring
and outdoor shading net provided
points the required
is shown in Figure environ-
2. In the
mental conditions
horizontal direction,for production.
there Tomatoes and
were nine measuring watermelons
points wereand
for temperature cultivated in the
relative humid-
ity in the greenhouse, and one outdoor measuring point was 20 m away from the green-
house. In the vertical direction, 3 levels were set according to the height of the cultivation
plots, the middle of the plant, and the top of the plant, and the outdoor measuring points
were set at the same level as shown in Figure 3. The temperature and relative humidity were
automatically recorded every 10 min. The location of the light intensity measuring point was
Agronomy 2024, 14, 876 3 of 21

experimental greenhouse in different seasons. The environmental factors such as light

intensity,
Figure indoor and
1. Experimental outdoor temperature, and the relative humidity of the greenhouse
greenhouse.
were tested using a nine-point measurement method.
The
The layout
layout of of
thethe indoor
indoor andand outdoor
outdoor measuring
measuring points
points is shown
is shown in Figure
in Figure 2. In 2.
theIn
the horizontal direction, there were nine measuring points for temperature
horizontal direction, there were nine measuring points for temperature and relative humid- and relative
ityhumidity in the greenhouse,
in the greenhouse, and onemeasuring
and one outdoor outdoor measuring
point was 20 point was from
m away 20 m the
away from
green-
the greenhouse.
house. In the verticalIndirection,
the vertical direction,
3 levels 3 levels
were set weretoset
according theaccording
height of to
thethe height of
cultivation
plots, the middle of the plant, and the top of the plant, and the outdoor measuring outdoor
the cultivation plots, the middle of the plant, and the top of the plant, and the points
measuring points were set at the same level as shown in Figure 3. The temperature and
were set at the same level as shown in Figure 3. The temperature and relative humidity were
relative humidity were automatically recorded every 10 min. The location of the light
automatically recorded every 10 min. The location of the light intensity measuring point was
intensity measuring point was the same as that of the temperature and humidity test
the same as that of the temperature and humidity test point, with a height of 2.0 m. Light
point, with a height of 2.0 m. Light intensity was measured every 5 min from 8:00 to 17:00
intensity was measured every 5 min from 8:00 to 17:00 every day.
every day.
Table 1 shows the environmental monitoring sensors used in this experiment and
Table 1 shows the environmental monitoring sensors used in this experiment and
their accuracy. The indoor and outdoor temperature, relative humidity, light intensity,
their accuracy. The indoor and outdoor temperature, relative humidity, light intensity, and
and ventilation rate were recorded as the thermal and light conditions.
ventilation rate were recorded as the thermal and light conditions.

Agronomy 2024, 14, x FOR PEER REVIEW 4 of 22

Figure 2. 2.
Figure Schematic of of
Schematic measuring points
measuring in in
points thethe
horizontal direction
horizontal (A,
direction B and
(A, CC
B and was
wasthethe
group serial
group serial
number of measuring points).
number of measuring points).

Figure3.3.Schematic
Figure Schematicof
ofmeasuring
measuringpoints
pointsin
inthe
thevertical
verticaldirection.
direction.

Table 1. Information of environment monitoring sensors.

Environment Parameter Instrument Model Precision

Air Temperature HOBO 0.2 °C
Agronomy 2024, 14, 876 4 of 21

Table 1. Information of environment monitoring sensors.

Environment Parameter Instrument Model Precision

Air Temperature HOBO 0.2 ◦ C
Relative Humidity HOBO ±2.5% Rh
Illumination Intensity Top Cloud-agri TPJ-22-G ±2% FS
Roof Temperature Fluke F59 ±2.0 ◦ C
Airflow Velocity Kestrel4000 0.10 m/s

2.2. Modeling and Numerical Simulation

2.2.1. Environment Parameters
The light intensity in the greenhouse was the average of the measurements from
8:00 to 17:00 during the daytime. The indoor temperature and relative humidity were
the averages of 27 groups of temperature and relative humidity. The outside temperature
and relative humidity were the averages of 3 groups measured outside the greenhouse.
In addition, the temperature and relative humidity of a certain layer in the greenhouse were
the average values of 9 groups at the layer. The light rate in the greenhouse was calculated
by Equation (1).
E
C = in (1)
Eout
where C denotes the light rate in the greenhouse, Ein denotes the average light intensity
inside the greenhouse, and Eout denotes the average light intensity outside the greenhouse.
The uniformity of light intensity in the greenhouse was calculated by Equation (2).

SD
CV = (2)
MN
where SD represents the standard deviation of light intensity in the greenhouse and MN
represents the average value of light intensity in the greenhouse.

2.2.2. Computational Domain and Meshes

The computational domain replicating the experimental setup was simulated using
ANSYS-Fluent (release 2016, ANSYS, Canonsburg, PA, USA). This study used the three-
dimensional (3D) steady-state method to solve the governing equations.
Agronomy 2024, 14, x FOR PEER REVIEW 5 of 22
As shown in Figure 4, this study established a 3D geometric model of the experimental
greenhouse, which took the southwest corner of the greenhouse as the coordinate origin
(0,0,0), the east–west direction as the transverse direction (span direction), the north–
north–south direction
south direction as the as the longitudinal
longitudinal direction
direction (depth(depth direction),
direction), the depth
the depth direction
direction of the
of
greenhouse as the y-axis positive direction, the span direction of the greenhouse as theasx-
the greenhouse as the y-axis positive direction, the span direction of the greenhouse
the
axisx-axis positive
positive direction,
direction, the height
the height direction
direction of the greenhouse
of the greenhouse as the
as the z-axis z-axis direction,
positive positive
direction, and the top window opening angle as
and the top window opening angle as a maximum of 45◦ .a maximum of 45°.

Figure4.
Figure Geometricmodel
4.Geometric modelof
ofgreenhouse.
greenhouse.

As shown in Figure 5, the greenhouse was located at the center of the external flow
field, and the ground center points of the two fields coincided. The length of the external
watershed was 10 times larger than the greenhouse building to ensure full development
of the fluid flow process [13]. The size of the inner flow field was 72 m (L) × 72 m (W) × 7.9
m (H), and the size of the outer flow field was 792 m (L) × 232 m (W) × 48 m (H) under
Agronomy 2024, 14, 876 Figure 4. Geometric model of greenhouse. 5 of 21

As shown in Figure 5, the greenhouse was located at the center of the external
field, and the
As shown in Figure 5,ground center points
the greenhouse of the two
was located fields
at the coincided.
center The length
of the external flowof the ext
field, and the ground center points of the two fields coincided. The length of the external wa-developm
watershed was 10 times larger than the greenhouse building to ensure full
tershed was 10 oftimes
the fluid flow
larger process
than [13]. The size
the greenhouse of thetoinner
building flow
ensure fullfield was 72 m (L)
development × 72 m (W)
of the
m (H),
fluid flow process and
[13]. Thethe size
size of the
of the outer
inner flowflow
fieldfield was
was 72 m792 m 72
(L) × (L)m× (W)
232 ×m 7.9
(W)m×(H),
48 m (H) u
and the size ofnatural ventilation.
the outer flow field was 792 m (L) × 232 m (W) × 48 m (H) under natu-
ral ventilation.

(a)

(b)
Figure
Figure 5. The front 5. The
view frontside
(a) and view (a) (b)
view andofside
the view (b) and
internal of the internal
external and external
airflow fields. airflow fields.

Considering
Considering the irregularitytheofirregularity of the top
the top structure structure
of the of the this
greenhouse, greenhouse,
study usedthis study
ICEM to divide the computational
ICEM domain into unstructured
to divide the computational domain intogrids, increase the
unstructured density
grids, of the de
increase
Agronomy 2024, 14, x FOR PEER REVIEW 6 of 22
the grid on the
ofvent and wall
the grid on theof vent
the greenhouse,
and wall of and carry out photocoagulation
the greenhouse, treatment.
and carry out photocoagulation t
After pre-experiments,
ment. the number of grids generated was 7,130,172, the number of
grid nodes was 1,162,296, the grid had no negative
After pre-experiments, volume,
the number and the
of grids grid quality
generated was greater
was 7,130,172, the numb
greater than 0.3.
than 0.3. As shown As shown
in Figure
grid nodes in
was 6, Figure 6, the grid
the grid division
1,162,296, division
the grid of theno
had of the greenhouse
greenhouse model under
negative volume, model under
natural
and the grid quality
natural ventilation
ventilation was determined.
was determined.

Figure6.
Figure 6. Global
Global grid
grid division
division of
ofgreenhouse
greenhousemodel.
model.

2.2.3.
2.2.3. Governing
Governing Equations
Equations
The
The airflow in the
airflow in the greenhouse
greenhouse isis aa low-speed
low-speed flow
flow field
field at
at room
roomtemperature,
temperature, which
which
can be regarded as an incompressible fluid [14], and its transport process
can be regarded as an incompressible fluid [14], and its transport process satisfies satisfies the
the gov-
governing equation in Equation
erning equation in Equation (3): (3):
∂(ρφ)
∂(ρφ )
+ div(ρφ
+ div (ρφU)U)
== (Γgradφφ)
div(Γgrad
div ) ++Sφ
Sφ (3)
(3)
∂t ∂t
In Equation (3), ρ is the fluid density, U is the velocity vector of (u, v, w), Γ is the
generalized diffusion coefficient, and Sφ is the source item.
The continuity equation is given by Equation (4) when φ = 1 , where Sm is the
source item of mass.
∂ρ ∂ρμi
Agronomy 2024, 14, 876 6 of 21

In Equation (3), ρ is the fluid density, U is the velocity vector of (u, v, w), Γ is the
generalized diffusion coefficient, and Sφ is the source item.
The continuity equation is given by Equation (4) when φ = 1, where Sm is the source
item of mass.
∂ρ ∂ρµi
+ = Sm (4)
∂t ∂xi
The momentum equation is given by Equations (5)–(7) when φ = U, where Su , Sv and
Sw are the source items of momentum.

∂(ρu) ∂ρ
+ div(ρφu) = div(Γgradu) + Su − (5)
∂t ∂x
∂(ρv) ∂ρ
+ div(ρφv) = div(Γgradv) + Sv − (6)
∂t ∂x
∂(ρw) ∂ρ
+ div(ρφw) = div(Γgradw) + Sw − (7)
∂t ∂x
The energy equation is given by Equation (8) when φ = T, where T is the temperature
of the airflow in the greenhouse and p is the air pressure in the greenhouse.
 
∂(ρE) ∂ ∂ ∂T

+ (ρµi E + pµi ) = Keff + µj τij eff + Sh − Hw (8)
∂t ∂xi ∂xi ∂xi

In Equation (8), Keff is effective thermal conductivity given by Equation (9), where Kt
is the turbulent heat transfer coefficient determined by the turbulence model.

Keff = K + Kt (9)

p µ2i
E = h− + (10)
ρ 2
p
h = ∑j mj Jj + (11)
ρ
In Equations (10) and (11), E is the total energy of the fluid, and h is the total enthalpy
of an ideal gas.
∂
∂xi ∑j j
Hw = hj J (12)

In Equation (12), Hw is the enthalpy of the humid air transport process. And hj is
given by Equation (13), where Teff is 298.15 K.
Z T
hj = Cpj dT (13)
Teff

2.2.4. Numerical Model

In this study, the heat transfers and airflow in the greenhouse under natural ventilation
were considered. The measured wind speed was less than 2.0 m/s, which accorded with the
low-speed flow field. The airflow process in the test area was fully developed, the air in the
simulated area can be regarded as a steady incompressible fluid, and the airflow conforms
to Boussinesq hypothesis. In addition, the flow state of the fluid in the greenhouse was
judged by the ratio of the Rayleigh number to the Prandtl number, which conformed to
the turbulent motion form [6,9,14]. The standard k-ε model was selected to simulate the
greenhouse under natural ventilation in Equations (14)–(16).

ρk2
µt = C µ (14)
ε
Agronomy 2024, 14, 876 7 of 21

 
∂ ∂ ∂ µt ∂k ∂uj ∂ui ∂uj
(ρk) + (ρkui ) = [ + µ1 ] + µt ( + ) − ρε (15)
∂t ∂xi ∂xj σk ∂xj ∂xi ∂xj ∂xi

ε2
 
∂ ∂ ∂ µt ∂ε ∂uj ∂ui ∂uj
(ρε) + (ρεui ) = [ + µ1 ] + C1 µt ( + ) − ( C2 ρ ) (16)
∂t ∂xi ∂xj σε ∂xj ∂xi ∂xj ∂xi k
In Equation (14), µt is the eddy viscosity coefficient, k is the turbulence fluctuation
kinetic energy, ε is the turbulence dissipation rate, and Cµ is the turbulence constant, while
in Equations (15) and (16), ui is the laminar eddy viscosity coefficient, and C1 , C2 , σk , σε are
empirical constants.
In addition, solar radiation is another important factor affecting the environmental
distribution in the greenhouse, and the influence of solar radiation on the microclimate in
the greenhouse is closely related to the solar azimuth angle and the geographical position
of the greenhouse. In this paper, discrete ordinates (DO) was the radiation model, and the
radiation transfer equation is shown in Equation (17).

σT 4
Z 4π
σs
∇( I (→, →) → +( a + σs ) I (→, →)) = an2 + I (→, →)ϕ(→, →, )dΩ′ (17)
r s s r s π 4π 0 r s s s

In Equation (17), Ω′ is the solid space angle of direct sunlight, which conforms to
Equation (18). 
cosα = (sinδ − sinβcosθ)/cosβsinθ
(18)
cosθ = sinβsinδ + cosβcosδcost
where α is the solar zenith angle, θ is the solar azimuth, β is the latitude of the test area, δ is
the solar declination, and t is the hour angle. In addition, sinδ is calculated by Equation (19).

sinδ = 0.39795cos(0.98563(N − 173)) (19)

In Equation (19), N indicates the number of days before January 1st of the current year.
→ → →
In Equation (17), r is the position vector, s is the direction vector, S′ is the scattering
direction, s is the length along the route, a is the absorption coefficient, n is the refractive
index, σs is the scattering coefficient, σ is the Stefan–Boltzmann constant, I is the radiation
→ →
intensity which depends on position r and direction s , T is the local temperature, and
K, ϕ are phase functions.
Solar Ray Tracing (SRT) was used to load the solar-load model to calculate the solar
radiation intensity, and the Solar Calculator (SC) was used to set the geographical position
(109◦ 50′ E, 18◦ 26′ N) and time zone (GMT + 8) of Lingshui area. The trend of the greenhouse
was north–south, the due north direction was y-axis positive (0,1,0), and the due east
direction was x-axis positive (1,0,0).

2.2.5. Boundary Conditions

Based on the actual physical structure, the distribution of temperature and airflow
in the plastic greenhouse with a semi-open roof under natural ventilation was mainly
considered. The cover material in the simulation model was 0.15 mm aging-resistant
polyethylene film, and the surrounding protective materials were insect-proof nets and
general polyethylene film. The physical properties of the greenhouse model are shown
in Table 2.
For the convenience of numerical simulation and considering the computer perfor-
mance, reasonable assumptions were made after integrating the actual situation of the
greenhouse. The temperature of the coating was uniformly distributed, and the heat trans-
fer coefficient was constant. Crops cultivated in the greenhouse model were all in the
seedling stage, and the transpiration was minor, ignoring the transpiration of crops in the
greenhouse. The evaporation of soil in the greenhouse was ignored. The heat exchange
through doors and windows in the greenhouse was not considered. Because of the large
size of the experimental greenhouse, the experimental area was a part of the whole green-
Agronomy 2024, 14, 876 8 of 21

house, so all four sides of the experimental greenhouse were regarded as contacting the
outside world.

Table 2. Material properties in greenhouse.

Physical Parameter Air Plastic Film Soil

Density (kg·m−3 ) 1.22 923 1900
Specific Heat Capacity (J · (kg · K)−1 ) 1006.43 2300 2200
Thermal Conductivity 0.0242 0.38 2.00
Absorption Coefficient 0.19 0.37 0.50
Scattering Coefficient 0.00 0.30 1.00
Refractive Index 1.00 1.92 -
Emissivity 0.86 0.80 0.90

As shown in Table 3, the boundary condition parameters of the experimental green-

house numerical simulation model were set.

Table 3. Boundary condition settings.

Parameter Setting Value

Windward Side Velocity Entrance 1.4 m/s
Leeward Side Pressure Outlet 101,325 Pa
Top Side Plastic --
Air Temperature 302.75 K

In addition, the protective materials around the greenhouse were an insect-proof net
and general polyethylene film. The physical characteristics of the insect-proof net are
shown in Table 4.

Table 4. Physical properties of insect-proof net.

Parameter Value
Diameter of the line 0.18 mm
Porosity 0.41
Mesh size 0.5 mm × 0.5 mm, 40-mesh
Permeability coefficient 8.26 × 10−10 m2
Resistance coefficient 1.89

In the numerical simulation model, the insect-proof net was regarded as a one-
dimensional porous medium, which was according to the Porous Jump boundary condition.
According to Equation (20), the permeability coefficient K of the insect-proof net was calcu-
lated, and the porosity εsw was obtained from Equation (21), where L denotes the mesh
size of the insect-proof net and d denotes the diameter of the insect-proof net line.

K = 3.44 × 10−9 εsw 1.6 (20)

εsw = (L − d)2 /L2 (21)

2.2.6. Solution of Discrete Differential Equation

In this paper, the differential equation was solved based on the finite volume method,
and its discretization format is shown in Table 5.
In the CFD module, the finite volume method was used to solve the computational
domain. And the pressure-based solver was used to solve incompressible fluid flow
problems in numerical simulation models. For solving discrete equations, a separate
method requires less computation time, while a coupled method consumes a lot of memory,
and SIMPLE algorithm is efficient in solving incompressible flow fields, so the SIMPLE
algorithm was selected.
 Agronomy 2024, 14, 876 9 of 21

Table 5. Selection of discrete scheme.

Pressure Energy Momentum

k Dispersion ε Dispersion
Dispersion Dispersion Dispersion
Mass–force Second-order Second-order First-order First-order
weighting upwind upwind upwind upwind

In this study, the default convergence standard of the continuity equation and momen-
tum equation in ANSYS was lower than 10−3 , and that of the energy equation was lower
than 10−5 . After pre-experiments, the convergence standards were determined as follows:
continuity and momentum equations were 10−5 , and the energy equation was 10−7 . The
empirical constants in Equations (14) and (16) were set as follows:

Cu = 0.09, C1 = 1.44, C2 = 1.92, σk = 1.0, and σε = 1.33

Agronomy 2024, 14, x FOR PEER REVIEW
3. Results
3.1. Environment Parameter Analysis
3.1.1. Time-Variation Difference of Light Intensity
Figure 7 shows the indoor and outdoor light intensity of four seasons. Due to the
influence of the external shading net, the indoor light intensity was much lower than the
outdoor light intensity. Apparently, the tropical indoor light intensity in the summer and
autumn was higher than that in the winter and spring, with the maximum average value of
(9.3 ± 1.1) × 103 Lux in the summer and the minimum average value of (4.1 ± 0.8) × 103
Lux in the spring. The minimum light rate calculated by Equation (1) appeared 10 oflowest
22 in
the spring at 12.4%, while the maximum light rate was 16.10% in the winter. The mean
coefficient of variation in the four seasons in the greenhouse was lower than 10.0%.

Figure 7. Indoor and outdoor light intensity of greenhouse in four seasons.

Figure 7. Indoor and outdoor light intensity of greenhouse in four seasons.

The daily maximum light intensity period appeared from 9:00 to 14:00, while the
minimum light intensity period appeared at 8:00 and 17:00. In addition, after 15:00 in the
summer and autumn, the indoor light intensity was lower than the light saturation point
 Agronomy 2024, 14, 876 10 of 21

The daily maximum light intensity period appeared from 9:00 to 14:00, while the
minimum light intensity period appeared at 8:00 and 17:00. In addition, after 15:00 in the
summer and autumn, the indoor light intensity was lower than the light saturation point
(LSP) of the crop. Therefore, in the actual production process, the operation time of the
external shading net should be adjusted, and the sunshade net could be closed after 15:00.

3.1.2. Temperature Variation of Greenhouse

As shown in Figure 8, the greenhouse indoor temperature varied with seasons. Dur-
ing the experiment, the minimum ambient temperature in the Lingshui area was 18.3 ◦ C
in the winter, while the annual average temperature was 25.1 ± 0.5 ◦ C. The maximum
Agronomy 2024, 14, x FOR PEER REVIEW
ambient temperature was 41.0 ◦ C in the summer, and the annual average temperature
11 of 22
was
31.9 ± 0.4 ◦ C. Therefore, open-field cultivation in the Lingshui area cannot meet the tem-
perature demand of horticultural crops, so it is necessary to carry out protected cultivation.

Figure 8. Indoor and outdoor average air temperature of greenhouse in four seasons.
Figure 8. Indoor and outdoor average air temperature of greenhouse in four seasons.

The average indoor

The average temperature
indoor in thein
temperature summer was 31.4
the summer was±31.4
0.2 °C, which
± 0.2 ◦ C, was
whichslightly
was slightly
lowerlower
than the
thanindoor air temperature.
the indoor DuringDuring
air temperature. the high-temperature period from
the high-temperature 10:00
period to 10:00
from
14:00toin14:00
the summer, the maximum temperature difference between inside
in the summer, the maximum temperature difference between inside and outside and outside
the greenhouse
the greenhousewas 2.7
was °C.2.7
The maximum
◦ C. The maximum indoorindoor
temperature in the summer
temperature was 39.3
in the summer °C,39.3 ◦ C,
was
which lagged one hour behind the outdoor peak temperature. It is
which lagged one hour behind the outdoor peak temperature. It is common to have common to have con-
vective rain during
convective rainthe hottest
during thesummer
hottesttime at noon
summer timeinattropical
noon inareas, which
tropical reduces
areas, which the
reduces
external temperature in a short time. In the autumn, the temperature in the
the external temperature in a short time. In the autumn, the temperature in the greenhouse greenhouse
was between
was between 24 °C24and◦ C25 °C,25
and and the
◦ C, andmaximum temperature
the maximum difference
temperature between
difference indoorsindoors
between
and outdoors
and outdoorswas 7.8was°C7.8 from 10:00 10:00
◦ C from to 14:00. The outdoor
to 14:00. temperature
The outdoor was obviously
temperature was obviously
higher than the
higher thanindoor temperature,
the indoor temperature,mainly because
mainly the cloudy
because weather
the cloudy and theand
weather shading
the shading
net blocked the sunlight and heat. The indoor temperature in the
net blocked the sunlight and heat. The indoor temperature in the winter and winter and spring was spring
between 24 °C and 26 °C, and the lowest indoor temperature in the winter was 19.6 °C,
which was 1.3 °C higher than the outdoor temperature. The indoor temperature difference
from 19:00 to 7:00 the next day was 8.6 °C and 4.5 °C in the summer and autumn, respec-
tively, and 6.0 °C and 4.4 °C in the winter and spring.
Figure 9 shows the temperature difference at three heights in the greenhouse. In the
 Agronomy 2024, 14, 876 11 of 21

was between 24 ◦ C and 26 ◦ C, and the lowest indoor temperature in the winter was
19.6 ◦ C, which was 1.3 ◦ C higher than the outdoor temperature. The indoor temperature
difference from 19:00 to 7:00 the next day was 8.6 ◦ C and 4.5 ◦ C in the summer and autumn,
respectively, and 6.0 ◦ C and 4.4 ◦ C in the winter and spring.
Figure 9 shows the temperature difference at three heights in the greenhouse. In the
summer, the temperature difference between the three heights was obvious. From 9:00
to 14:00 in the summer, the minimum average temperature appeared at 3.5 m, while the
maximum average temperature appeared at 2.0 m. The maximum temperature difference
Agronomy 2024, 14, x FOR PEER REVIEW between these heights was 5.4 ◦ C, which was mainly related to the solar radiation
12 of 22 at noon,
the roof vent, and the transpiration rate of the cultivated plants. In addition, the average
temperature in the winter and spring increased with height and fluctuated less than 3 ◦ C.
This was mainly due to gentle temperature difference between indoors and outdoors in the
the winter and spring. It can be seen that with the increase in temperature in the green-
winter and spring. It can be seen that with the increase in temperature in the greenhouse,
house, the temperature difference between different layers was more significant.
the temperature difference between different layers was more significant.

Figure 9. Temperature variation of different heights (0.7 m, 2.0 m, 3.5 m) in four seasons.
Figure 9. Temperature variation of different heights (0.7 m, 2.0 m, 3.5 m) in four seasons.

3.1.3. Humidity
3.1.3. Humidity Variation
Variation of Greenhouse
of Greenhouse
As shownAs shown
in Table in6,Table 6, the range
the range of average
of average relative
relative humidity
humidity in theingreenhouse
the greenhousewaswas from
from 76%76%toto 87%,
87%, and
and thethe outdoor
outdoor average
average relative
relative humidity
humidity waswas
fromfrom
66%66% to 78%.
to 78%. Influenced
Influ-
enced bybythe
thecovering
coveringofof thethe greenhouse
greenhouse andand
the the transpiration
transpiration of plants,
of plants, the indoor
the indoor rela- relative
humidity
tive humidity waswas basically
basically higherhigher
than than the outdoor
the outdoor all yearallround.
year round.
Therefore,Therefore,
the highestthe highest
relativerelative
humidity humidity in the greenhouse
in the greenhouse generallygenerally
appeared appeared
at night at
or night
dawn.orAlong
dawn. Along
with the with the
rise of indoor
rise of indoor temperature
temperature during theduring the daytime,
daytime, the relative
the relative humidityhumidity in the greenhouse
in the greenhouse
dropped dropped to thefrom
to the lowest lowest from
11:00 11:00The
to 14:00. to 14:00. Thein
difference difference in relative
relative humidity humidity
indoors and indoors
outdoorsand outdoors
varied varied with
with seasons, and theseasons,
maximum and the maximum
difference difference
was during was and
summer during
au- summer
tumn. The maximum relative humidity difference between indoors and outdoors was
18.6% in the autumn, when there is the “dry season” of tropical monsoon climate with
less rainfall and dry air. From 10:00 to 16:00 in the summer, the outdoor average temper-
ature was 38.7 ± 0.2 °C, and the average relative humidity was 48.4 ± 0.8%, while the aver-
age temperature inside the greenhouse was 37.8 ± 0.1 °C and the average humidity was
 Agronomy 2024, 14, 876 12 of 21

and autumn. The maximum relative humidity difference between indoors and outdoors
Agronomy 2024, 14, x FOR PEER REVIEW 13 of 22
was 18.6% in the autumn, when there is the “dry season” of tropical monsoon climate
with less rainfall and dry air. From 10:00 to 16:00 in the summer, the outdoor average
temperature was 38.7 ± 0.2 ◦ C, and the average relative humidity was 48.4 ± 0.8%, while
Table 6. Relative humidity
the average of the greenhouse
temperature inside in
thefour was 37.8 ± 0.1 ◦ C and the average humidity
seasons (0:00–24:00).
greenhouse
was 54.5 ± 0.7%. The
Maximum (%)relative humidity
Minimum was(%)suitable for the plants,
Average ± SE but
(%) the temperature
Season was relatively
Indoor high, causing a humid
Outdoor Indoorand hot indoor environment.
Outdoor Indoor Outdoor
Spring 99.0 98.0 58.3 50.3 86.5 ± 0.2 77.9 ± 0.9
Table 6. Relative humidity of the greenhouse in four seasons (0:00–24:00).
Summer 97.1 96.1 40.2 40.6 76.6 ± 0.6 71.2 ± 2.2
Autumn 92.0 91.3
Maximum (%) 42.9 26.0
Minimum 79.0
(%) ± 0.2 66.4Average
± 1.0 ± SE (%)
Season
Winter 98.4 98.4
Indoor 46.3
Outdoor 43.7
Indoor 82.7 ±
Outdoor 0.3 73.5 ± 1.0
Indoor Outdoor
Spring 99.0 98.0 58.3 50.3 86.5 ± 0.2 77.9 ± 0.9
As shown in Figure 10,97.1
Summer the relative humidity
96.1 in the
40.2greenhouse40.6 was affected76.6by sea- 71.2 ± 2.2
± 0.6
Autumn
son and height. 92.0
Due to the influence 91.3 radiation 42.9
of solar and roof vents, 26.0the average ± 0.2
79.0 indoor 66.4 ± 1.0
relative humidity decreased with the increase in height in the spring, autumn, and winter, 73.5 ± 1.0
Winter 98.4 98.4 46.3 43.7 82.7 ± 0.3

and the indoor relative humidity changed in a gradient. There were great differences in
As shown
relative humidity amongindifferent
Figure 10, the relative
heights from 9:00 humidity
to 14:00ininthe
thegreenhouse
summer and was affected by season
winter.
and height.
In the summer, Due to the
the maximum influence
relative humidity of solar radiation
difference betweenand3.5roof vents,
m and thewas
2.0 m average indoor
31.4%, and the maximum
relative humidity relative humidity
decreased withdifference
the increasebetween 3.5 min
in height and
the0.7 m wasautumn,
spring, 23.7%. and winter,
In the winter,
and thethe indoor
difference in relative
relative humidity
humidity between
changed in a3.5 m and 0.7There
gradient. m is obvious,
were great anddifferences in
the maximum
relative difference
humidity in among
relative different
humidityheights
was 15.5%.fromAt9:00
night, the average
to 14:00 in the relative
summer and winter.
humidityInamong differentthe
the summer, heights was similar.
maximum relative humidity difference between 3.5 m and 2.0 m was
From the above
31.4%, and the results,
maximumit can relative
be statedhumidity
that solardifference
radiation between
and natural 3.5 ventilation
m and 0.7 m was 23.7%.
cause theInrise
theinwinter,
the indoor temperature
the difference inand the decrease
relative humidity in between
the relative3.5humidity
m and 0.7 during
m is obvious, and
the daytime,
the maximum difference in relative humidity was 15.5%. At night, the the
and it can be concluded that there is a coupling relationship between average relative
three environmental
humidity among factorsdifferent
of light intensity,
heights was air temperature,
similar. and relative humidity.

Figure 10.Figure
Relative
10.humidity
Relativevariation
humidity ofvariation
different heights (0.7 m,
of different 2.0 m,(0.7
heights 3.5 m)
m, in
2.0different sea-
m, 3.5 m) in different seasons.
sons.
 Agronomy 2024, 14, 876 13 of 21

From the above results, it can be stated that solar radiation and natural ventilation
cause the rise in the indoor temperature and the decrease in the relative humidity during
Agronomy 2024, 14, x FOR PEER REVIEWthe daytime, and it can be concluded that there is a coupling relationship between
14 of 22the three
environmental factors of light intensity, air temperature, and relative humidity.

3.2. Validation Studies on CFD Simulation in Greenhouse

3.2. Validation Studies on CFD Simulation in Greenhouse
3.2.1. Validation of CFD Simulation Model
3.2.1. Validation of CFD Simulation Model
The simulated temperature values at the same position as the experimental measure-
The simulated
ment points weretemperature
extractedvalues
fromatthe
thesimulation
same position as theand
results experimental
comparedmeasure-
with the actual
ment points were extracted from the simulation results and
measured temperature values. By comparison, it was found that the compared withspatial
the actual
distribution
measured temperature values. By comparison, it was found that the spatial distribution
law of simulated and measured values was consistent, but the simulated values of each
law of simulated and measured values was consistent, but the simulated values of each
measuring point were lower than the measured values. The maximum error between
measuring point were lower than the measured values. The maximum error between sim-
simulated and measured values appears at the test point C3 at 3.5 m height, with an error
ulated and◦ measured values appears at the test point C3 at 3.5 m height, with an error of
of 1.7 C and a relative error of 5.9%. However, the measured temperature values of the
1.7 °C and a relative error of 5.9%. However, the measured temperature values of the
whole measurement point were in agreement with the simulated values.
whole measurement point were in agreement with the simulated values.
As
As shownshown in Table
in Table 7, differences
7, the the differences between
between the measured
the measured and simulated
and simulated temperature
tempera-
ture at three heights in the greenhouse were compared, and the average relative error waswas 3.0%.
at three heights in the greenhouse were compared, and the average relative error
The
3.0%. error
The between
error betweenthe themeasured
measured values andsimulated
values and simulatedvalues
values in in
thethe engineering
engineering test test was
wasconsidered
considered acceptable. Therefore,the
acceptable. Therefore, theestablished
established greenhouse
greenhouse CFD CFD model
model is effective, the
is effective,
the boundary conditionsare
boundary conditions areset
setcorrectly,
correctly,and
andthethe model
model cancan be used
be used to further
to further analyze the
analyze
the distribution lawof
distribution law ofthe
thegreenhouse
greenhouseenvironment
environment and
and optimize
optimize thethe parameters.
parameters.

Table
Table 7. Comparison
7. Comparison between
between measured
measured and simulated
and simulated averageaverage temperature
temperature at three heights.
at three heights.

MeasuredMeasured
AverageAverage Simulated Average
Simulated Average
HeightHeight
(m) (m) Temperature (◦ C)Temperature
Temperature ◦ Relative Error (%)
Relative Error (%)
Temperature (°C) (°C) ( C)
0.7 0.7 26.5 26.5 26.2 26.2 2.5 2.5
2.0 27.0 26.3 2.8
2.0 3.5
27.0 27.5
26.3 26.4
2.8 3.9
3.5 27.5 26.4 3.9
3.2.2. Simulation Analysis of Greenhouse
3.2.2. Simulation Analysis of Greenhouse
Figure 11 shows the velocity vector diagram of three sections along the depth direction
Figure 11 shows the velocity vector diagram of three sections along the depth direc-
of the greenhouse: 12, 36, and 60 m. It can be seen that with the increase in the horizontal
tion of the greenhouse: 12, 36, and 60 m. It can be seen that with the increase in the hori-
distance of of
zontal distance thethe
greenhouse
greenhouseininthe thevertical
vertical direction, theairflow
direction, the airflowrate
rate
in in
thethe greenhouse
green-
house gradually weakened. The airflow rate in the cultivation area was slow, which was was not
gradually weakened. The airflow rate in the cultivation area was slow, which
notconducive
conducive to to the
theindoor
indoorand and outdoor
outdoor air exchange.
air exchange. This distribution
This distribution easily acaused a
easily caused
high-temperature zone in the greenhouse, which affected the growth
high-temperature zone in the greenhouse, which affected the growth and development and development
of of
cultivation.
cultivation. Atsame
At the the same
time, time, the airflow
the airflow at the at thewas
inlet inlet was blocked
blocked by the
by the film film
in the in the upper
upper
area,
area, which
which produced
produced a vortex
a vortex and increased
and increased the speed
the wind wind speed at the inlet.
at the inlet.

Figure
Figure 11. Velocity
11. Velocity on X on X sections
sections (Xm,
(X = 12 =X12=m,
36 X
m,=X36 m,m).
= 60 X = 60 m).
Agronomy 2024, 14, x FOR PEER REVIEW 15 of 22
Agronomy 2024, 14, x FOR PEER REVIEW 15 of 22
Agronomy 2024, 14, 876 14 of 21

Due to the tunnel ventilation, the vertical velocity component of the air outside the
greenhouse
Due after
Duetotothe entering
thetunnel
tunnel the greenhouse
ventilation,
ventilation,the was velocity
thevertical
verticalminor,
velocityas shown in Figure
component
component ofofthe 12.
theair Inoutside
addition,
airoutside the
the
the enclosure
greenhouse material
after was
entering thean insect-proof
greenhouse wasnet, so
minor, the
as indoor
shown and
in outdoor
Figure
greenhouse after entering the greenhouse was minor, as shown in Figure 12. In addition,12. temperature
In addition,
difference
the wasmaterial
theenclosure
enclosure small,
materialthe
was hot-pressing
was effectnet,
ananinsect-proof
insect-proof was not
net,so obvious,
sothe
theindoor and
indoorand the
and effect of
outdoor
outdoor the natural
temperature
temperature
ventilation
difference
differencewaswas
was weak.the
small,
small, thehot-pressing
hot-pressingeffect
effectwas
wasnotnotobvious,
obvious,and
andthe
theeffect
effectofofthe
thenatural
natural
ventilation
ventilationwaswasweak.
weak.

Figure 12. Velocity on Y sections (Y = 12 m, Y = 36 m, Y = 60 m).

Figure
Figure 12.12. Velocity
Velocity onon
YY sections
sections (Y(Y = 12
= 12 m,m,
YY = 36
= 36 m,m,
Y =Y 60
= 60
m).m).
Figure 13 shows the temperature contour of three sections along the depth direction
Figure 13 shows the temperature contour of three sections along the depth direction
of the greenhouse:
Figure 13 shows12, 36,temperature
the and 60 m. The temperature
contour changes along
of three sections in thethe
three sections
depth were
direction
of the greenhouse: 12, 36, and 60 m. The temperature changes in the three sections were
similar, the ventilation of the greenhouse is mainly limited by the external
of the greenhouse: 12, 36, and 60 m. The temperature changes in the three sections were wind speed,
similar, the ventilation of the greenhouse is mainly limited by the external wind speed, and
and thethe
similar, indoor temperature
ventilation of thevaried with the
greenhouse is length.
mainlyBecause
limited of
bythetheeffect of solar
external windradiation,
speed,
the indoor temperature varied with the length. Because of the effect of solar radiation, the
the the
and indoor temperature
indoor increased
temperature varied with height. In addition,
the length. thethe
Because of temperature difference
effect of solar be-
radiation,
indoor temperature increased with height. In addition, the temperature difference between
tween
the different
indoor heightsincreased
temperature of the greenhouse wasInwithin
with height. 0.5
addition, °C,
the and the indoor
temperature temperature
difference be-
different heights of the greenhouse was within 0.5 ◦ C, and the indoor temperature field
field was
tween relatively
different stable.
heights
was relatively stable. of the greenhouse was within 0.5 °C, and the indoor temperature
field was relatively stable.

Figure
Figure13.
13.Temperature
Temperaturecontour
contouron
onXXsections
sections(X
(X==12
12m,
m,XX==3636m,
m,XX==6060m).
m).
Figure 13. Temperature contour on X sections (X = 12 m, X = 36 m, X = 60 m).
3.3.Optimization
3.3. OptimizationSimulation
SimulationofofGreenhouse
Greenhouse
Accordingto
3.3. Optimization
According tothe
thesimulation
simulation
Simulation analysis,the
of Greenhouse
analysis, thecombined
combinedventilation
ventilationeffect
effectofoftunnel
tunnelven-
ven-
tilation and roof ventilation was obvious. Simulation results showed that the ventilation
tilation and roof
According toventilation was analysis,
the simulation obvious. the
Simulation
combinedresults showed
ventilation that of
effect thetunnel
ventilation
ven-
in the greenhouse was mainly caused by wind pressure, and the temperature was greatly
in the greenhouse was mainly caused by wind pressure, and the temperature was
tilation and roof ventilation was obvious. Simulation results showed that the ventilation greatly
affected by external wind speed. To solve this problem, the temperature and flow field
inaffected by external
the greenhouse waswind speed.
mainly To solve
caused this pressure,
by wind problem, and
the temperature and was
the temperature flowgreatly
field in
in the greenhouse were analyzed by changing parameters such as the configuration and
the greenhouse
affected were
by external analyzed
wind by solve
speed. To changing parameters
this problem, thesuch as the configuration
temperature and flow fieldand in
opening size of the windows and the external wind speed, which would affect the natural
the greenhouse were analyzed by changing parameters such as the configuration and
ventilation of the greenhouse. Therefore, the established CFD model of the greenhouse
Agronomy 2024, 14, x FOR PEER REVIEW 16 of 22
Agronomy 2024, 14, x FOR PEER REVIEW 16 of 22

Agronomy 2024, 14, 876 opening size of the windows and the external wind speed, which would affect the natural15 of 21
opening
ventilation of size of the windows
the greenhouse. and the the
Therefore, external wind speed,
established which would
CFD model affect the natural
of the greenhouse
ventilation of the greenhouse. Therefore, the established
was used to analyze the influence of the above parameters on the natural ventilation CFD model of the greenhouse
of
was used
the greenhouse, to analyze
hopefully the
to influence
provide a of the
theoreticalabove
basis parameters
for the on the
environmental natural ventilation
control and of
was used to analyze the influence of the above parameters on the natural ventilation of
the greenhouse,
optimization hopefully
of large-scale film to provide a theoretical
greenhouses in tropicalbasisareas.for the environmental control and
the greenhouse, hopefully to provide a theoretical basis for the environmental control and
optimization of large-scale film greenhouses
optimization of large-scale film greenhouses in tropical areas. in tropical areas.
3.3.1. Greenhouse Ventilation Simulation
3.3.1.
3.3.1.
The Greenhouse
film of theVentilation
Greenhouse
roof windwardSimulation
Ventilation Simulation
surface of the greenhouse was modified to an insect-
proof net to The roof
roof film
increase
The theof
film the
the windward
ventilation
of area, surface
windward and the of
surface ofthe
other greenhouse
thesurfaces
greenhouse was
wasmodified
remained the same,
modified to an
aninsect-
towith insect-
proof
the previous net
proof net to increase
settings. In the
to increase the
the ventilation
improved
ventilation area,
model, and
area, the the
andchanges other
the other surfaces
insurfaces
temperatureremained the
and flow
remained same,
field with
the same, with
the
in thethe previous
greenhouse settings. In the improved model, the changes
previous were analyzed. improved model, the changes in temperature and flow in temperature and flow field
field in
in
thethe
As greenhouse
greenhouse
shown were
in Figure were analyzed.
analyzed.
14, with the increased vents, turbulence disappeared in the
greenhouse. As
AsInshown
shown in
in Figure
Figure 14,
the longitudinal 14, with
with the
direction the increased
increased
of the greenhouse,vents,
vents,theturbulence
maximumdisappeared
turbulence difference inin
disappeared in the
the
greenhouse.
greenhouse.
flow velocity betweenInInthe 0.7longitudinal
the longitudinal
m, 2.0 m, and direction
direction
7.5 m was of of
the the
aboutgreenhouse,
0.05 m/s, the
greenhouse, and maximum
the airflowdifference
maximum
the difference
distri- in
flow
in velocity
flow velocitybetween
between 0.7 m,
0.7 2.0
m, m,
2.0and
m, 7.5
and m was
7.5 m about
was 0.05
about
bution in the greenhouse was relatively stable, with an average velocity of 1.3 m/s. In the m/s,
0.05 and
m/s,the airflow
and the distri-
airflow
bution
Y direction in the
distribution
of thein greenhouse
the greenhouse
greenhouse, was relatively
wasairflow
the top stable,
relatively with
stable,
velocity an
theaverage
inwith an velocity
average
second of 1.3
halfvelocity
decreased of m/s.
1.3 In the
m/s.
and In
Ythe
direction
was mainly Y direction
affectedof the greenhouse,
of the
by the length of the
greenhouse, top
thethe airflow
top airflow
greenhouse. velocity
velocity ininthethesecond
secondhalf
halfdecreased
decreasedand and
was
was mainly
mainly affected
affected by by the
the length
length ofof the
the greenhouse.
greenhouse.

Figure 14. Vector diagram of airflow in greenhouse with increased vents on X sections (X = 12 m, X
Figure
= 36 m, X = 6014.
Figure m).Vector
14. Vectordiagram
diagramofofairflow
airflowiningreenhouse
greenhousewith
withincreased vents
increased onon
vents X sections (X (X
X sections = 12 m, m,
= 12 X
=X36 m, m,
= 36 X =X60= m).
60 m).
Figure 15 shows the temperature in the greenhouse was evenly distributed in the
FigureThe
Figure
lateral direction. 15 shows
15 shows the greenhouse
top of the temperaturewas
temperature in the
in the greenhouse
greenhouse
affected wassolar
was
by external evenly
evenly distributed
distributed
radiation in the
in thein the
lateral
lateral
vertical direction.
direction.
direction; The top
The top
therefore, of the greenhouse
theoftemperature
the greenhouse was
was
in the affected by external
affected byincreases
greenhouse solar
external with radiation
solarheight. in
radiation in thethe
vertical direction; therefore,
vertical direction; therefore, thethe temperature in the greenhouse increases with height.

Figure 15. Temperature in contour greenhouse with increased vents on X sections (X = 12 m, X = 36 m,

X = 60 m).
Agronomy 2024, 14, x FOR PEER REVIEW 17 of 22
Agronomy 2024, 14, x FOR PEER REVIEW 17 of 22

Figure 15. Temperature in contour greenhouse with increased vents on X sections (X = 12 m, X = 36

Agronomy 2024, 14, 876 Figure 15. Temperature in contour greenhouse with increased vents on X sections (X = 12 m, X16=of3621
m, X = 60 m).
m, X = 60 m).
3.3.2. Greenhouse Ventilation Optimizations
3.3.2.
3.3.2.Greenhouse
GreenhouseVentilation
VentilationOptimizations
Optimizations
Figure 16 shows the flow field distribution of the combined ventilation of side and
Figure 16 shows
Figure 16Byshows the flow
the flow field distribution ofofthe combined ventilation ofofside and
roof windows. increasing the field
open distribution
area of the side the combined
windows, the ventilation
overall airflowside and
of the
roof
roofwindows.
windows. ByByincreasing
increasing the open
the area
open of
area the
of side
the windows,
side windows, the overall
the airflow
overall of
airflow theof
greenhouse was improved. The flow field distributions of Y = 12 m, Y = 36 m, and Y = 60
greenhouse was improved. The flow field
flowdistributions of Y = of
12 Ym,=Y12= m,
36 m,
Y =and Y =and
60
mthe greenhouse
sections was improved.
in the greenhouse wereThe
basically field distributions
the same, and the average velocity of36airflow
m,
mY sections
= 60 in the greenhouse
m sectionswas in the were basically the same, and the average velocity of airflow
in the greenhouse 0.6 greenhouse were basically
m/s. In addition, the increase theinsame, and the
the open areaaverage velocity
of the side win-of
inairflow
the greenhouse
in the was 0.6 m/s.
greenhouse was In addition,
0.6 m/s. In the increase
addition, inincrease
the the openinareathe of the area
open side of
win-
the
dows both strengthened the cross ventilation and affected the thermal pressure ventila-
dows
side both strengthened
windows both the cross the
strengthened ventilation
cross and affected
ventilation and the thermal
affected thepressure
thermal ventila-
pressure
tion. Therefore, the air velocity near the top window reached from 0.5 m/s to 1 m/s.
tion. Therefore,
ventilation. the air velocity
Therefore, near thenear
the air velocity topthe
window reached
top window from 0.5
reached fromm/s0.5
tom/s
1 m/s.to 1 m/s.

Figure 16.16.
Vector diagram
Vector ofof
diagram airflow distribution
airflow in in
distribution greenhouse with
greenhouse side
with window
side (Y(Y
window = 12,12,
YY = 36,36,
YY
Figure
Figure
= 60).
16. Vector diagram of airflow distribution in greenhouse with side window (Y ==12, Y ==36,
Y ==60).
60).

AsAsshown
shownin Figure
Figure17,17,in the
thedirection
directionparallel
paralleltotothe
thegreenhouse,
greenhouse,the thetemperature
As shown ininFigure 17, ininthe direction parallel to the greenhouse, the temperature
temperature
distribution of the
distributionofofthe three
thethreesections
threesections
sections had certain
had differences.
certain Under
differences. the
Under influence of solar ra-
distribution had certain differences. Under the the influence
influence of solar
of solar ra-
diation, the temperature change increases with the increase in height in the
radiation, the temperature change increases with the increase in height in the longitudinal longitudinal
diation, the temperature change increases with the increase in height in the longitudinal
direction, and
direction, and thethe
temperature
temperature difference at Xat= X
difference 12=section in theintransverse
12 section direction
the transverse was
direction
direction, and the temperature difference at X = 12 section in the transverse direction was
large.
was large.
large.

Figure
Figure 17.17. Temperature
Temperature contour
contour inin greenhouse
greenhouse with
with side
side window(X(X= =
window 1212
m,m,
XX = 36
= 36 m,m,
X =X 60
= 60
m).m).
Figure 17. Temperature contour in greenhouse with side window (X = 12 m, X = 36 m, X = 60 m).
3.3.3. External Airflow Simulation
By comparing the experiment and simulation results, it can be concluded that the ven-
tilation efficiency of the greenhouse was mainly affected by the wind pressure ventilation,
 3.3.3. External Airflow Simulation
3.3.3.
By External
comparingAirflow Simulationand simulation results, it can be concluded that the
the experiment
Agronomy 2024, 14, 876 ventilation efficiency of
By comparing the
the greenhouse
experiment wassimulation
and mainly affected
results,byit the
canwind pressure venti-
be concluded that
17 the
of 21
lation, so increasing
ventilation the of
efficiency outdoor ventilation
the greenhouse rate
was could affected
mainly improveby thethe
airflow
wind of the green-
pressure venti-
house. Figure
lation, 18 showsthe
so increasing theoutdoor
indoor airflow vector
ventilation ratediagram for the outdoor
could improve wind
the airflow ofspeed of
the green-
3 m/s.
house. Figure 18 shows the indoor airflow vector diagram for the outdoor wind
so increasing the outdoor ventilation rate could improve the airflow of the greenhouse. speed of
3Figure
m/s. 18 shows the indoor airflow vector diagram for the outdoor wind speed of 3 m/s.

Figure 18. Vector diagram of external wind speed at 3 m/s (X = 12 m, X = 36 m, X = 60 m).

Figure18.
Figure Vectordiagram
18.Vector diagramof
ofexternal
externalwind
windspeed
speedatat33m/s
m/s(X(X==1212m,
m,XX= =3636m,m,X X= =
6060 m).
m).
With the increase in greenhouse length, the speed gradually decreased, so the venti-
With
lation inWith the increase
the increasewas
the greenhouse in greenhouse
in greenhouse length,
mainly affected
length,bythethespeed
the gradually
external
speed decreased,
wind speed.
gradually Whenso
decreased, the
the
so the ventila-
wind
venti-
tion
speed
lation in
was the greenhouse
3 m/s,
in the the indoor
greenhouse was mainly
wasairmainly affected
velocity by
increased,
affected the
by theand external wind
there wind
external was an speed. When
obvious
speed. When the
gradientwind
the wind
speed
change was 3 m/s, the indoor air velocity increased, and there was an obvious gradient
speedinwas the 3vertical
m/s, the direction,
indoor where the average
air velocity increased,velocity
and of the airflow
there was an at the height
obvious of
gradient
change
2.0change
m in the in the vertical direction, where the average velocity of the airflow at the height of
in greenhouse was the maximum,
the vertical direction, where thereaching
average 2.1 m/s. The
velocity of theairflow
airflow from theheight
at the gutter of
2.0 m in the greenhouse was the maximum, reaching 2.1 m/s. The airflow from the gutter
to 2.0
the mroof
in was evenly distributed
the greenhouse was theunder the influence
maximum, reaching of2.1
greenhouse covers. from
m/s. The airflow Therefore, in
the gutter
to the roof was evenly distributed under the influence of greenhouse covers. Therefore, in
a tropical
to the roof area,
wasitevenly
is recommended
distributed to design
under the the greenhouse
influence according
of greenhouse to the
covers. summer in
Therefore,
a tropical area, it is recommended to design the greenhouse according to the summer wind
wind direction
a tropical to make
area, full use of the natural
it is recommended to design ventilation.
the greenhouse according to the summer
direction to make full use of the natural ventilation.
As shown in Figure 19, when the external wind
wind direction to make full use of the natural ventilation. speed was 3 m/s, the indoor temper-
As shown in Figure 19, when the external wind speed was 3 m/s, the indoor tem-
ature distribution
As shown in trend was19,
Figure consistent
when theinexternal
the lateralwinddirection,
speed was and3them/s,temperature grad-
the indoor temper-
perature distribution trend was consistent in the lateral direction, and the temperature
ually increased
ature in thetrend
distribution vertical
wasdirection,
consistent which
in thewas mainly
lateral affected
direction, bythe
and solar radiation and
temperature grad-
gradually increased in the vertical direction, which was mainly affected by solar radiation
airflow
ually speed
increasedin the
in greenhouse. At the same
the vertical direction, time,
which themainly
was plasticaffected
film on the greenhouse
by solar radiation roof
and
and airflow speed in the greenhouse. At the same time, the plastic film on the greenhouse
played an speed
airflow obstructing
in the role, which made
greenhouse. At thethe temperature
same in the top
time, the plastic filmarea at the
on the entrance of
greenhouse roof
roof played an obstructing role, which made the temperature in the top area at the entrance
theplayed
vent slightly higher. role, which made the temperature in the top area at the entrance of
an obstructing
of the vent slightly higher.
the vent slightly higher.

Figure 19. Temperature contour in the greenhouse with external wind speed at 3 m/s (X = 12 m,
X = 36 m, X = 60 m).

3.3.4. Optimization Schemes

By comparing the optimization modes of enlarging the size of roof vents, increasing
side windows, and changing the external wind speed, the indoor air velocity gradually
increased with the external wind speed, which had an obvious turbulent effect on the
Agronomy 2024, 14, 876 18 of 21

indoor air velocity at different heights. When the size of greenhouse vents was increased,
the air velocity at the height of 2.0 m was the maximum, the relative velocity at both
sides was small, and the overall air velocity in the greenhouse was relatively uniform. In
addition, changing the configuration of vents also had a certain effect on the distribution of
air velocity in the greenhouse. By increasing the open area of side windows, the utilization
rate of roof vents would be improved, and the ventilation rate in the greenhouse could
be increased.
Therefore, the spatial distribution of microclimate in the greenhouse is affected by the
external wind speed, the configuration of vents, and the size of vents. The optimization
scheme of designing the orientation of the greenhouse according to the maximum air
volume in the summer can improve the utilization efficiency of natural ventilation, and
increasing the size of greenhouse vents and adding side windows can be selectively used
in the winter and spring.

4. Discussion
A plastic greenhouse with a semi-open roof is mainly used to produce truss toma-
toes and mini watermelons in tropical areas. Therefore, this study mainly analyzes the
annual production environment of this greenhouse. Compared with the gothic arch green-
house [15] and Venlo-type greenhouse [16], a plastic greenhouse with a semi-open roof
exhibits a high natural ventilation efficiency and lower cooling energy consumption in the
summer. A semi-open ventilation structure, which increases the air exchange rate of the
greenhouse, is adopted for the roof, and the north–south trend improves the overall light
uniformity, which is beneficial to annual production.
The light intensity in the plastic greenhouse was significantly lower than the outdoor
natural light due to the influence of roof angle, frame structure, plastic film characteristics,
and other factors. Due to the influence of the shading net, the indoor light conditions of
four seasons cannot meet the growth needs, which is different from the research results of
Singh et al. [17] and is mainly related to the cleanliness of the greenhouse structure and
covering materials. At the same time, the solar radiation received by the greenhouse in
the south–north direction was lower than that in the east–west direction, but the sunlight
received by the greenhouse was more uniform throughout the year [18].
The indoor temperature is strongly affected by solar radiation and the ambient air
temperature, and the highest indoor temperature during the experiment reached 39.3 ◦ C
in the summer, lagging behind the outdoor temperature by one hour, which is similar
to the results of Badji et al. [19]. The indoor temperature is still higher than the suitable
temperature for the crops in the greenhouse, so additional ventilation and cooling measures
are needed. The relative humidity in the test greenhouse was lower than that of the solar
greenhouse and plastic greenhouse because of the large area of the side windows and
the top windows with a long time of opening. However, the relative humidity in the test
greenhouse changed slowly, which was due to the larger volume of the test greenhouse
and higher humidity inertia.
In this study, different numbers and types of grids were selected for comparative
analysis in the process of meshing. Pre-experiments showed that it was faster to transform
unstructured grids into polyhedral grids for calculation within the applicable scope of
computers, which was in agreement with the research results of Banakar et al. [20]. In
addition, for the selection of boundary conditions in the process of simulating natural
ventilation in the greenhouse, two sides were set as symmetry boundary conditions in this
paper, which is different from the periodic boundary conditions set by Deng et al. [21],
mainly because of the differences between the actual conditions of the two physical models,
which cannot be considered equally. Furthermore, the thermal environment simulation
error of the numerical simulation model constructed in this study was 3.0%, which is far less
than the allowable range of 20% in engineering simulations. This result is similar to those
of Liang et al. [22], in which the maximum relative error of temperature simulation was
3.71%. Compared with the dynamic simulation model, the three-dimensional steady-state
Agronomy 2024, 14, 876 19 of 21

model used in this study provides a lower computational load, and its accuracy is close to
that of Liang and Xu et al. [23].
According to the optimization simulation results of the greenhouse CFD model, opti-
mizing the size and configuration of vents can promote roof ventilation efficiency, which
is in agreement with the results of Villagran et al. [6] on natural ventilation of traditional
Colombian greenhouses, gothic multi-span greenhouses, and curved multi-span green-
houses in tropical areas. In addition, the optimization scheme proposed in this study
is superior to the results of Singhal et al. [24] and Fan et al. [25] in terms of large-scale
environmental uniformity and natural ventilation efficiency. At the same time, when the
wind speed outside the greenhouse is low, the influence of wind direction on natural
ventilation is as important as that of wind speed, which is consistent with the research
results of Bournet et al. Moreover, compared with the results of Lu et al. [26], who used
multifunctional fan-coil unit systems in Chinese solar greenhouses, this study had higher
energy efficiency by increasing the vent size and adopting the top window, which improved
the uniformity of the thermal environment in the greenhouse and the efficiency of natural
ventilation without additional mechanical equipment.

5. Conclusions
This study evaluated the environmental parameters of a plastic greenhouse with a
semi-open roof in a tropical area in four seasons and optimized the vent structure of the
greenhouse based on CFD simulation. The following conclusions were drawn:
1. In this study, the temperature, relative humidity, and light intensity in different
seasons were analyzed for a plastic greenhouse with a semi-open roof. The maximum
indoor temperature in the summer was 39.3 ◦ C, and the average indoor temperature
in the summer was 31.4 ± 0.2 ◦ C, while the average indoor temperature was between
24 ◦ C and 26 ◦ C in the other three seasons; the minimum indoor temperature in the
winter was 19.6 ◦ C. The average relative humidity was 76–87% annually. The average
indoor light intensity varied less than 10.0% in four seasons.
2. The three-dimensional steady-state CFD model of a multi-span plastic greenhouse
under natural ventilation was established. The maximum relative error (MRE) of
the model between the simulated and measured values of the temperature at the
measuring point in the greenhouse was 5.9%, and the average relative error was 3.0%.
3. Increasing side windows and roof vents can make full use of natural wind and achieve
a better energy-saving effect by reducing the usage of mechanical ventilation systems.
With the increase in vent size, the indoor airflow distribution was stable, and the
average velocity airflow was 1.3 m/s.
4. The indoor airflow was homogenized and improved by increasing the side window
area, after which the average velocity of indoor air was 1.4 m/s, and the ventilation
efficiency of the roof window was improved. When the inlet wind speed of the
greenhouse was increased to 3.0 m/s, the average velocity at the cultivation area
reached 2.1 m/s. Furthermore, it is recommended that future studies include different
types of plants within the CFD model since the developed plants will generate thermal
changes of different airflow patterns from those found in this study.

Author Contributions: H.Y.: conceptualization, methodology, formal analysis, writing—original

draft, writing—review and editing, visualization. K.W.: software, writing—review and editing. H.Y.
and K.W. contributed equally to this work. J.Z.: software, data curation, resources. Z.P.: methodology,
project administration, writing—review and editing, Funding acquisition. All authors have read and
agreed to the published version of the manuscript.
Funding: This research was funded by the Hainan Province Science and Technology Special Fund
(ZDYF2023XDNY044).
Informed Consent Statement: This study was performed by the authors, and no participants other
than the authors were involved in it; informed consent was obtained from all authors.
Agronomy 2024, 14, 876 20 of 21

Data Availability Statement: Data will be made available on request.

Acknowledgments: The author would like to thank all the anonymous reviewers for their helpful
comments and suggestions and Geqi Yan B.E. for help and contribution to the field experiment.
Conflicts of Interest: The authors declare no conflicts of interest.

Nomenclature

C Daylighting rate in greenhouse Sφ Source item

CFD Computational fluid dynamics φ Universal variable
CV Coefficient of variation Sm Source item of mass
Ein Average illumination intensity inside the greenhouse T Temperature of airflow
Eout Average illumination intensity outside the greenhouse p Air pressure
SD Standard deviation Keff Effective thermal conductivity
MN Average value of light intensity Kt Turbulent heat transfer coefficient
SRT Solar Ray Tracing E Total energy of fluid
ρ Fluid density h Total enthalpy of ideal gas
U Velocity vector of (u, v, w) Hw Enthalpy of humid air transport process
Γ Generalized diffusion coefficient Teff Reference temperature
µt Eddy viscosity coefficient k Turbulence fluctuation kinetic energy
ε Turbulence dissipation rate Cµ Turbulence constant
ui Laminar eddy viscosity coefficient C1 , C2 , σk , σε Empirical constants
Ω, Solid space angle of direct sunlight α Solar zenith angle
θ Solar azimuth β Latitude of the test area
δ Solar declination t Hour angle
→
N The number of days before January 1st of thecurrent year r Position vector
→ →
s Direction vector S′ Scattering direction
s Length along the route a Absorption coefficient
n Refractive index σs Scattering coefficient
σ Stefan–Boltzmann constant I Radiation intensity
K Permeability coefficient K, ϕ Phase functions
L Mesh size of insect-proof net εsw Porosity of insect-proof net
d Diameter of the insect-proof net line

References
1. Chen, W.H.; Mattson, N.S.; You, F. Intelligent control and energy optimization in controlled environment agriculture via nonlinear
model predictive control of semi-closed greenhouse. Appl. Energy 2022, 320, 119334. [CrossRef]
2. Ma, D.; Carpenter, N.; Maki, H.; Rehman, T.U.; Tuinstra, M.R.; Jin, J. Greenhouse environment modeling and simulation for
microclimate control. Comput. Electron. Agric. 2019, 162, 134–142. [CrossRef]
3. Li, A.; Huang, L.; Zhang, T. Field test and analysis of microclimate in naturally ventilated single-sloped greenhouses. Energy
Build. 2017, 138, 479–489. [CrossRef]
4. Bay, E.; Martinez-Molina, A.; Dupont, W.A. Assessment of natural ventilation strategies in historical buildings in a hot and humid
climate using energy and CFD simulations. J. Build. Eng. 2022, 51, 104287. [CrossRef]
5. Lee, S.Y.; Lee, I.B.; Kim, R.W. Evaluation of wind-driven natural ventilation of single-span greenhouses built on reclaimed coastal
land. Biosyst. Eng. 2018, 171, 120–142. [CrossRef]
6. Villagran, E.A.; Romero, E.J.B.; Bojacá, C.R. Transient CFD analysis of the natural ventilation of three types of greenhouses used
for agricultural production in a tropical mountain climate. Biosyst. Eng. 2019, 188, 288–304. [CrossRef]
7. Soussi, M.; Chaibi, M.T.; Buchholz, M.; Saghrouni, Z. Comprehensive review on climate control and cooling systems in
greenhouses under hot and arid conditions. Agronomy 2022, 12, 626. [CrossRef]
8. Tawalbeh, M.; Aljaghoub, H.; Alami, A.H.; Olabi, A.G. Selection criteria of cooling technologies for sustainable greenhouses:
A comprehensive review. Therm. Sci. Eng. Prog. 2013, 38, 101666. [CrossRef]
9. Wu, X.; Li, Y.; Jiang, L.; Wang, Y.; Liu, X.; Li, T. A systematic analysis of multiple structural parameters of Chinese solar greenhouse
based on the thermal performance. Energy 2023, 273, 127193. [CrossRef]
10. Liu, J.; Wu, X.; Sun, F.; Wang, B. Development and Application of a Crossed Multi-Arch Greenhouse in Tropical China. Agriculture
2022, 12, 2164. [CrossRef]
11. McCartney, L.; Orsat, V.; Lefsrud, M.G. An experimental study of the cooling performance and airflow patterns in a model
Natural Ventilation Augmented Cooling (NVAC) greenhouse. Biosyst. Eng. 2018, 174, 173–189. [CrossRef]
Agronomy 2024, 14, 876 21 of 21

12. Flores-Velázquez, J.; Rojano, F.; Aguilar-Rodríguez, C.E.; Villagran, E.; Villarreal-Guerrero, F. Greenhouse Thermal Effectiveness
to Produce Tomatoes Assessed by a Temperature-Based Index. Agronomy 2022, 12, 1158. [CrossRef]
13. Wang, X.W.; Luo, J.Y.; Li, X.P. CFD based study of heterogeneous microclimate in a typical Chinese greenhouse in central China.
J. Integr. Agric. 2013, 12, 914–923. [CrossRef]
14. Paing, S.T.; Anderson, T.N. Characterizing the effects of geometry on the natural convection heat transfer in closed even-span
gable-roof greenhouses. Therm. Sci. Eng. Prog. 2024, 48, 102408. [CrossRef]
15. Hashemi, S.F.; Pourfallah, M.; Gholinia, M. Thermal performance enhancement in an indirect solar greenhouse dryer using helical
fin under variable solar irradiation. Sol. Energy 2024, 267, 112217. [CrossRef]
16. Yan, H.; Deng, S.; Zhang, C.; Wang, G.; Zhao, S.; Li, M.; Zhou, Y. Determination of energy partition of a cucumber grown
Venlo-type greenhouse in southeast China. Agric. Water Manag. 2023, 276, 108047. [CrossRef]
17. Singh, M.C.; Kaur, P.; Singh, J.P.; Singh, K.G.; Singh, A. Postharvest Physicochemical Quality Assessment of Greenhouse Produced
Soilless Cucumbers in Relation to Differential Fertigation. J. Agric. Eng. 2023, 60, 38–50.
18. Bo, Y.; Zhang, Y.; Zheng, K.; Zhang, J.; Wang, X.; Sun, J.; Wang, J. Light environment simulation for a three-span plastic greenhouse
based on greenhouse light environment simulation software. Energy 2023, 271, 126966. [CrossRef]
19. Badji, A.; Benseddik, A.; Bensaha, H.; Boukhelifa, A.; Hasrane, I. Design, technology, and management of greenhouse: A review.
J. Clean. Prod. 2022, 373, 133753. [CrossRef]
20. Banakar, A.; Montazeri, M.; Ghobadian, B.; Pasdarshahri, H.; Kamrani, F. Energy analysis and assessing heating and cooling
demands of closed greenhouse in Iran. Therm. Sci. Eng. Prog. 2021, 25, 101042. [CrossRef]
21. Deng, L.; Huang, L.; Zhang, Y.; Li, A.; Gao, R.; Zhang, L.; Lei, W. Analytic model for calculation of soil temperature and heat
balance of bare soil surface in solar greenhouse. Sol. Energy 2023, 249, 312–326. [CrossRef]
22. Liang, Z.; Xu, L.; De Baerdemaeker, J.; Li, Y.; Saeys, W. Optimization of a multi-duct cleaning device for rice combine harvesters
utilizing CFD and experiments. Biosyst. Eng. 2020, 190, 25–40. [CrossRef]
23. Xu, T.; Zhou, H.; Lv, X.; Lei, X.; Tao, S. Study of the distribution characteristics of the airflow field in tree canopies based on the
CFD model. Agronomy 2022, 12, 3072. [CrossRef]
24. Singhal, S.; Yadav, A.K.; Prakash, R. Numerical modelling of an earth-air-tunnel assisted single span saw-tooth greenhouse for
tropical climate. Int. J. Therm. Sci. 2023, 187, 108138. [CrossRef]
25. Fan, Z.; Li, Y.; Jiang, L.; Wang, L.; Li, T.; Liu, X. Analysis of the Effect of Exhaust Configuration and Shape Parameters of
Ventilation Windows on Microclimate in Round Arch Solar Greenhouse. Sustainability 2023, 15, 6432. [CrossRef]
26. Lu, J.; Li, H.; He, X.; Zong, C.; Song, W.; Zhao, S. CFD Simulation and Uniformity Optimization of the Airflow Field in Chinese
Solar Greenhouses Using the Multifunctional Fan–Coil Unit System. Agronomy 2023, 13, 197. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.