A Two-Factor Thermal Screen Control Strategy for Chinese Solar Greenhouses in High-Latitude Areas

Abstract

Covering thermal screen on the front roof is one of the most general methods to improve the thermal performance of the solar greenhouse in China. Thermal screen control, however, is operator-dependent and based on empirical strategies. In order to more effectively manage the thermal screen, an optimal control method based on solar radiation and temperature difference between indoor and outdoor was established. The influence of different factors on the control of greenhouse thermal screen is systematically analyzed and the control function of the greenhouse thermal screen was calculated. The empirical control formula was established based on simulation which lasted for 62 days. As a result, the two-factor control method can significantly improve the air temperature when the thermal screen is controlled, and it can increase the average air temperature by 0.53 °C. Comparing with temperature difference, solar radiation has a greater impact on the control of thermal screen. The control method based on temperature difference and solar radiation can save 7.2% energy in winter. The research can provide reference for energy saving and automatic control of Chinese solar greenhouse.

1. Introduction

Solar greenhouses, as a type of protected agricultural building, are widely used in the production of vegetables, fruits and flowers worldwide [1]. They use solar radiation as an energy source, which can effectively reduce energy consumption and gas emissions from the greenhouses [2]. The Chinese solar greenhouse (CSG) is the main crop facility production construction in some cold countries like China, Korea and Canada (Figure 1a). CSG is a type of agricultural production facility that can realize the off-season production of crops at low operating costs. During a frigid winter, a CSG can provide 30 °C temperature differences between the greenhouse and outside environment [3]. In 2018, the total CSG area in China has reached over 570,000 hectares, accounting for 59.8% of the greenhouse area in China “http://www.sheshiyuanyi.com/ (accessed on 15 January 2023)”. The distribution of CSG is summarized in Figure 2. CSG is the major production facility of vegetables in China, especially at higher latitudes and low temperature areas. CSG not only fills the winter vegetable gap in the northern area of China, but also increases farmers’ income.

The typical construction of the CSG consists the north wall, east wall, west wall, back roof and front roof, as shown in Figure 1 [4,5,6]. A research has shown that the front roof causes most of the heat loss in a CSG and heat loss can reach 70% of the overall greenhouse in extreme weather [7]. Therefore, increasing the insulating ability of the front roof can observably improve the thermal performance of the CSG, especially during the cold season [8]. To reduce heat loss on the front roof of the greenhouse, thermal screens (thermal insulation quilt) are widely used in CSGs from October to May. Thermal screens are high thermal-resistance materials with low density and soft texture, such as grass curtains and cotton quilts. During a cold night, a thermal screen is closed to prevent massive heat loss and the thermal screen is opened in the daytime so that the sunlight can warm the greenhouse through the front roof. Therefore, research on the thermal screen can improve thermal performance of a greenhouse at night [9]. Many studies have proven that thermal screens can effectively improve the thermal performance of greenhouses at night. The thermal conductivity of the thermal screen can observably affect the heating requirements of a greenhouse during the cold season [10]. The contributions of different thermal screen materials and different thermal screen combinations to the thermal performance of greenhouses were systematically studied during a typical winter [11]. In addition, many studies have shown that the control method of thermal insulation quilt also significantly affects the thermal performance of the greenhouse. Wang et al. (2014) have shown that the opening and closing time of the insulation quilt can significantly affect the air temperature of the greenhouse in the morning and at night [12]. Zhang et al. (2020) have shown that the management of thermal screen is vital for the growth of tomato plants [13]. Gilli et al. discovered that a proper thermal screen management method can save energy without affecting crop quality [14]. Therefore, it is essential to obtain an ideal thermal screen management strategy to improve the performance of CSG.

The control method of thermal screens involves opening and closing them at a scheduled time in China. The control method of thermal-screen management relies on the experience of the operator, especially on cloudy days] and this control method extremely depends on the judgment of the managers [14,15,16] so as to avoid misjudgments on the opening and closing times of thermal screens, which would subsequently lead to a reduction in thermal performance. Therefore, a large number of studies have explored the reasonable management of thermal screens. A study calculated the regularity of thermal screens in different months, which is based on local sunrise and sunset times [13]. Another study explored the influence of heat-preservation control methods on greenhouse temperatures and the means to generate a control strategy accordingly [17]. In addition, a control method based on the outdoor solar radiation intensity was proven to be a feasible method [18]. As can be seen from the above research, temperature and solar radiation are the keys which affect the control mode of thermal screens. However, the above studies only considered a single influencing factor and did not consider the combined effect of the two factors on the greenhouse thermal screen control strategy. The omission of an influencing factor will lead to a decrease in accuracy. In addition, due to the aggravation of rural population aging and the continuous loss of labor force, the development of rural labor force will accelerate before 2035 [19]. The mechanization and automation of agricultural production will inevitably replace the labor-oriented production mode, which will become the development trend of facility horticulture in the future. Therefore, there is an urgent need for the automatic control of solar greenhouse thermal insulation control method. The study of control strategies can reduce the consumption of labor and increase the output value of greenhouse.

In order to obtain a dynamic thermal screen control method, which is based on indoor-outdoor temperature. In this paper, an air temperature prediction model and an energy flow calculation model of a solar greenhouse in Shenyang were established. The influence of indoor and outdoor temperature differences and external solar radiation on the reasonable opening and closing times of greenhouse thermal screen were analyzed. According to these rules, the rule of indoor and outdoor temperature difference, as well as the illumination in a solar greenhouse when the thermal screen is turned on and off during a typical meteorological month, are calculated, then, the control strategy is established and the effectiveness of the two-factor control method was also evaluated. This study can provide a theoretical basis for the opening and closing of thermal screens in cold regions in China and provide a method and a consultation for the automatic control of thermal screens.

2. Materials and Methods

2.1. Energy Balance Calculation

In a greenhouse environment simulation, the key to calculating the opening and closing time of the thermal screen is to calculate the energy balance of the greenhouse. According to the heat balance rule, the control method of thermal screens should ensure that the greenhouse receives more energy. In other words, the thermal screen should be opened when the front roof is able to compensate for the heat lost from the greenhouse. At night, the thermal screen is turned off when the greenhouse accumulator stores the most heat. Therefore, the calculation of energy gains and energy loss in the greenhouse is very important in this paper.

CSG, as a kind of passive solar energy harvesting building, acquires energy mainly from sunlight [20]. Therefore, the solar radiation should be calculated. The solar trajectory changes with time, and the azimuth and altitude of the sun changes from hour to hour. These factors lead to a difference in solar radiation energy gained per hour [21]. Thus, for greenhouses, solar altitude is the key to elevating the gain energy. The solar altitude can be calculated by Equation (1). The solar azimuth angle θ can be calculated by Equation (2) as

$$\sinh =\sin \delta ·\sin \phi +\cos \delta ·\cos \phi ·\cos \eta$$

(1)

$$\cos \theta =\frac{(\sinh ·\sin \phi -\sin \delta )}{\cosh ·\cos \phi }$$

(2)

where φ is the local latitude of the greenhouse, δ is the solar declination and η is the solar hour angle.

The outside direct solar radiation on normal surface ($J_h$) can be calculated by Equation (3) as

$${{\displaystyle J}}_h={{\displaystyle J}}_n·\cos (90°-\theta )$$

(3)

where $J_n$ is the amount of direct solar radiation through the atmosphere.

$${{\displaystyle J}}_n=({{\displaystyle J}}_0/r^2)p^{\mathrm{csch}}$$

(4)

$$\mathrm{r}=r_n/r_0$$

(5)

where P is the coefficient of atmospheric transparency in one day, J₀ is the solar radiation constant 1353 w/m², $r_n$ is the distance from the Earth to the sun in one day, $r_0$ is the average distance from the Earth to the sun in one year.

However, the greenhouse energy gain is also closely related to diffused solar radiation. $J_s$ can be represented as

$${{\displaystyle J}}_s=1.2·{{\displaystyle J}}_h·(1-p^{\mathrm{csch}})/(1-1.4\ln p)$$

(6)

The greenhouse gaining solar radiation energy of the front roof can be represented as

$${{\displaystyle J}}_r=({{\displaystyle J}}_h+{{\displaystyle J}}_s)A_f\tau$$

(7)

where τ is the solar radiation transmittance of the front roof and $A_f$ is the front roof area.

The gain energy of the greenhouse can be considered as the heat accumulated from opening and closing thermal screens. Thus, in this greenhouse, the final gain energy in the greenhouse can be presented as

$${{\displaystyle Q}}_{in}={\displaystyle \sum _{open}^{close}{{\displaystyle J}}_r}$$

(8)

during winter in Shenyang, the thermal screen usually opened at 8:30 and closed at 15:30, which resulted in a total of 8 h of daylight.

The loss of greenhouse energy is another important index that affects greenhouse performance, especially in winter. The greenhouse daily energy loss can be determined as

$${{\displaystyle Q}}_{out}={\displaystyle \sum _{0:00}^{24:00}({{\displaystyle U}}_{evn}}+{{\displaystyle U}}_{inf}+{{\displaystyle U}}_{pla}+{{\displaystyle U}}_{act}+{{\displaystyle U}}_{ven})$$

(9)

where $U_{evn}$ is the heat loss from the greenhouse envelope and soil, $U_{inf}$ is the heat loss due to air infiltration, $U_{pla}$ is the heat consumption by plants due to water evaporation in the greenhouse, $U_{act}$ is the energy generated by human activities and lighting, and $U_{ven}$ is the heat loss due to greenhouse ventilation in daytime.

Since the greenhouse has less ventilation in winter, it is assumed that the greenhouse is a closed space and the heat loss caused by ventilation can be ignored. In this study, the air infiltration was set at 0.35 per hour. The energy produced by human activity, lighting equipment and electronic equipment in greenhouses is much lower than the energy produced by sunlight, so these factors can be ignored. In greenhouses, at least 70 percent of the solar energy absorbed by plants is converted into heat in the air [22], so the influence of plants is negligible. Therefore, the energy loss in one day can be expressed as

$${{\displaystyle Q}}_{out}={\displaystyle \sum _{0:00}^{24:00}({{\displaystyle U}}_{evn}}+{{\displaystyle U}}_{inf})$$

(10)

The enclosure structure of the greenhouse can be divided into walls (including side wall, east wall, west wall and back slope), front roof and soil. Therefore, U₁ can be calculated by

$$U_1=U_{wo}+U_{fo}+U_{so}$$

(11)

where $U_{wo}$ is the heat loss from the outside surface of the wall, $U_{fo}$ is the heat loss from the front roof and $U_{so}$ is the heat conduction to the deep soil.

Not all of the energy captured in the greenhouse is released into the air, but some of it is stored in the walls and soil and released at night and on cloudy days to maintain the temperature of the greenhouse at night. Therefore, this part of the energy needs to be calculated separately, the heat stored by a greenhouse in one period ($Q_g$) can be expressed as

$$Q_g={\displaystyle \sum _{\mathrm{start}}^{\mathrm{end}}(S_w+S_s)}$$

(12)

$$S_w=U_{wo}+U_{wi}$$

(13)

$$S_{\mathrm{s}}=U_{so}+U_{si}$$

(14)

where $S_w$ is the energy stored in walls, $S_s$ is the energy stored in shallow soil and $U_{wi}$ is the is the heat loss from the inside surface of the walls.

The heat load of a heating system refers to the amount of heat that the heating system supplies to the building in unit time to reach the required indoor temperature. This index is one of the important indexes to evaluate the energy-saving performance of buildings. Therefore, heat load is taken as the evaluation index of control method in this paper. The calculation method of heat load can be obtained as

$$Q_h=Q_u+Q_v$$

(15)

where $Q_u$ is the heat lost when the greenhouse air reaches setting temperature, and $Q_v$ is the heat energy gained by the greenhouse. At night, the heat source in the greenhouse can be regarded as the soil and the internal surface of the wall, and the heat dissipation of the greenhouse is mainly through the front roof, when the air temperature is lower than the temperature of the soil and the wall. During continuous overcast days or extreme conditions, the heat accumulated in greenhouse walls often only supports heating for a period of time. Therefore, the greenhouse wall will become a heat dissipation structure when the air temperature is higher than the wall surface temperature.

$$Q_u={\displaystyle \sum {\mathrm{A}}_{\mathrm{i}}{K}_i(t_{set}-t_{out})}+{\rho }_{air}{c}_{air}L(t_{set}-t_{out})$$

(16)

$$\mathrm{L}=NV$$

(17)

where ${\mathrm{A}}_{\mathrm{i}}$ the area of one structure, $K_i$ is the heat transfer coefficient of one structure, $t_{set}$ is the setting air temperature, $t_{out}$ is the outside air temperature, ${\rho }_{air}$ is the density of air, $c_{air}$ specific heat at constant pressure, $c_{air}$ = 1.003 (kJ/kg·°C), L is the infiltrate air volume, $N$ is air change coefficient (0.35 per hour) and $V$ is the volume of the greenhouse.

2.2. Reference Greenhouse

The solar greenhouse used in this study is located in Shenyang, Liaoning Province (north latitude of 41°48′, east longitude 123°25′), which has a temperate monsoon climate, and it is always cold and dry in winter. As a typical Liaoshen second generation energy-saving solar greenhouse (Figure 1a), the experimental greenhouse has a length of 60.0 m, width of 8.5 m and height of 4.5 m. The north wall of this greenhouse has a height of 3.0 m, and the azimuth angle of the greenhouse is 6° from south to west to obtain more light during the daytime. The front roof of the greenhouse is covered by polyolefin (PO) film, which allows light to heat the greenhouse during daytime, and there is a thermal screen on the top of the front roof which can roll down and prevent internal heat loss from the front roof during the night. The thermal screen is constructed of multilevel cotton blankets with a thickness of 0.06 m. The structure of the greenhouse is shown in Figure 1b.

2.3. Experimental Arrangement

To prove the accuracy of the model, real weather data were employed to compare the simulated temperature with the measured temperature. The experiment was conducted in December 2020, and the weather conditions were monitored from 0:00 on 19 December 2020 to 24:00 on 22 December 2020, for 96 h. The outdoor air temperature and solar radiation were collected by the outdoor weather station located in the experimental base with an accuracy of ±0.1 °C and 20 W/m², and the indoor air temperature was collected by temperature and humidity sensors (RC4) with an accuracy of ±0.1 °C. There were nine sensors in the experimental greenhouse and they were divided into three groups, where each group had three sensors. The arrangement form of these sensors in each group is shown in Figure 1c. Three groups of measuring points were, respectively, arranged 15 m away from the east and west walls of the greenhouse, so as to measure the average air temperature of the greenhouse.

2.4. Calculation Model

To analyze the energy balance, creating a greenhouse dynamic modeling is important. The solar greenhouse geometry model in this study was built in SketchUp, and the thermal simulation was carried out by EnergyPlus. EnergyPlus is widely used in building thermal environment simulations and energy consumption analyses [16]. In recent years, it has also been widely used in agricultural building research [23,24,25]. In this model, the front roof is divided into three rectangles, and the shading of the greenhouse skeleton is ignored. The PO films are considered as single-glass windows, and the effect of the greenhouse wall waterproof material on the greenhouse is ignored. The heat balance algorithm in this study is the conduction transfer function (CTF), which is a sensible heat calculation method that does not take into account moisture storage or diffusion in the construction elements. The time step of this study was set to six times per hour. In this model, the heat exchanges of the greenhouse were carefully considered, including walls, shallow soil, PO films and air, ignoring the effects of plants and humidity on energy balance.

The north wall of the solar greenhouse is one of its most important parts because the north wall stores solar energy during the day and releases it into the air at night to maintain night time temperatures. The north wall of the solar greenhouse in this study is made of 370 mm brick blocks and 130 mm thick insulation board. The brick wall is located inside the greenhouse and the insulation board is located outside the greenhouse. The front roof of the greenhouse is covered by PO films during the day and a thermal screen at night. The thickness of the PO films and thermal screen are 0.2 mm and 50 mm, respectively, and the thermal characteristics of the soil at a depth of 500 mm are also considered in this model. The model also considers the influence of infiltration on the greenhouse heat balance, and the air change coefficient is set at 0.35 per hour. Due to the low temperature of the greenhouse in winter, there is little natural ventilation, so the impact is ignored. The physical parameters of the materials in this model are listed in

Table 1. Physical parameters of materials.

Material	Block Brick	PO Films	Soil	Insulation Board (XPS)	Thermal Screen
Thickness (mm)	370	0.2	500	130	50
Conductivity (w/m·k)	0.81	0.64	0.45	0.03	0.1
Density (kg/m3)	1800	--	1100	50	--
Specific heat(J/kg·K)	1050	--	600	1380	--

.

To verify the accuracy of this model, the greenhouse air temperature was measured on a typical sunny day. Figure 3 shows the greenhouse air temperature and simulation air temperature. Measurements over the whole period were used to compare with simulated values. The error analysis method in this paper adopts the following equation [26]:

$$IA=1-\frac{{\displaystyle {\sum }_{y=1}^N(X_{py}-X_{my})}}{{\displaystyle {\sum }_{y=1}^N(|X_{py}-X_{pave}|}+|X_{py}-X_{pave}|{)}^2}$$

(18)

where $X_{py}$ and $X_{my}$ are numerical and experimental values, $X_{pave}$ and $X_{mave}$ are numerical and experimental average values.

In this error analysis equation, the range of IA can change between 0 and 1. When IA is 1, the model is proved to be completely consistent with the measuring value. On the contrary, when the value of IA is 0, it means that the model value is completely inconsistent with the experimental value (numerical and experimental results are not matching each other). In this case, IA is 0.9908, which indicates that this model has a high level of accuracy. The model can be used in the greenhouse indoor temperature calculation simulations.

3. Results

3.1. Calculation of Opening Time

To analyze the effect of solar radiation on opening and closing thermal screens, a real typical sunny and a real typical cloudy day were monitored in December (solar radiation and middle temperature). To simulate the effect of temperature, virtual outdoor air temperatures were established on a sunny and a cloudy day, as shown in Figure 4.

According to the theory of heat transfer rules in buildings, the environmental factors that affect greenhouse energy gain are the outdoor direct solar radiation rate and the outdoor diffuse solar radiation rate. These factors are often easy to measure by a meteorological station. However, there are many factors affecting the lost energy in solar greenhouses, in addition to the air temperatures inside and outside the greenhouse, wind speed, wind direction, relative humidity, effective temperature of the sky and many other factors. These factors often need professional equipment to measure or calculate, and they undoubtedly increase the cost and difficulty of monitoring. Therefore, to simplify the control model of thermal screens, it is important to establish a relationship model between indoor and outdoor air temperature differences and solar radiation rate when thermal screens are opened and closed [27]. In this study, typical sunny days and typical cloudy days in Shenyang winter were used to change their external temperatures, and the time for the front roof of the greenhouse to reach thermal equilibrium was calculated.

According to the heat balance rules, the thermal screen should be opened rapidly, when the energy gained by the front roof of the greenhouse is more than the energy lost [16]. Figure 5 shows the influence of different external temperatures on the heat balance of the front roof of the greenhouse under sunny and cloudy conditions. The weather data used in this model are shown in Figure 4. In this case, the middle temperature is a typical sunny and cloudy day in winter, and the low temperature and high temperature are the climatic conditions that both reduce and increase air temperature by 10 °C, respectively. As seen from Figure 5a, the average heat loss of the roof under low temperature increased by 4.20 MJ, while the high temperature decreased by 4.09 MJ compared with the middle temperature from 8:00–9:00. On the time scale, low external temperature on a clear day causes a 10 min delay in opening the thermal screen, while high external temperature in general causes the opening of the thermal screen to be earlier by 10 min. Figure 5b shows the influence of different greenhouse external temperatures on the heat balance of the front roof of the greenhouse in low light. From 9:00 to 10:00, the average roof heat loss before low temperature increased by 4.25 MJ compared to the middle temperature, while the high temperature decreased by 4.43 MJ. The effect of temperature on the uncovering time is similar to that on sunny days. At the same time, comparing Figure 5a with Figure 5b, it can be found that overcast days can delay the opening of the greenhouse thermal screen by one hour.

In order to obtain a reasonable control method of thermal insulation quilt, typical cold meteorological months in Shenyang were used to calculate the greenhouse heat preservation at different opening times (Figure 6). In this case, 61 days in December and January are considered in the simulation.

In order to obtain the empirical function of opening thermal screen, the temperature difference between indoor and outdoor air and the solar radiation outside the greenhouse are recorded under typical months and last for 62 days. The indoor-outdoor air temperature difference and solar radiation were recorded when the thermal screen was properly opening. Figure 7 shows the relationship of inside-outside air temperature difference and solar radiation at the optimal opening time of the thermal screen. It can be seen that they have a strong association, as shown in the figure, and the fitting formula is y = 0.0918x² − 0.1186x + 80.943. R² is 0.8187, proving that the function fits the calculation results. The formula has a high accuracy, which can be used to predict the opening time of the thermal screen.

3.2. Calculation of Closing Time

In a solar greenhouse, the solar radiation energy is not directly released into the air, but the greenhouse walls and soil surface absorb the light and convert it into heat, part of which is transferred into the deep wall and soil, and the other part is lost through radiation and convection. A large amount of study has proved the significance of heat storage on greenhouse night air temperature [3,5,28].

Therefore, the key which determines the night performance of a greenhouse is the heat accumulated of heat storage material. If the thermal screen closed too early, the wall and the soil will not accumulate the most energy, and if the thermal screen closed late, more energy will be lost from the greenhouse. Thus, the thermal screen should be closed when the greenhouse has the most heat reserves.

Figure 8 shows the influence of different greenhouse external temperatures on the heat storage of greenhouse under sunny and cloudy conditions. In this case, the heat storage of greenhouse walls and soil is calculated from 15:00 to 17:00. The weather data for simulation is also shown in Figure 4. As shown in Figure 8a, on sunny days, each 10 °C rising in outdoor temperature increases the heat storage of the heat storage structures by 50 MJ. Under low, medium and high temperatures, the opening time of the thermal screen is 15:10, 15:20 and 15:30, respectively. As shown in Figure 8b, on cloudy days, each 10 °C increase in outdoor temperature increases the heat storage of the wall by 57 MJ. Under low, medium and high temperatures, the closing time of the thermal screen is 15:40, 16:00 and 16:00, and the change pattern is similar to that of sunny days. Compared with sunny days, due to the greenhouse temperature and heat storage material drop caused by cloudy days, the greenhouse heat accumulator accumulates more energy on cloudy days. Therefore, cloudy days lead to a delay in the closing of greenhouse thermal screen.

To determine the relationship between the solar radiation rate and the indoor-outdoor temperature difference when the greenhouse thermal screen is closed, this section adopts a similar research approach as the previous section. Figure 9 shows the temperature difference and outdoor solar radiation at the ideal closing time of the thermal screen in December and January. It can be seen that they have a functional relationship, and the fitting formula is y = 0.1781x² − 1.7721x + 95.606. R² is 0.8028. The formula has a high fitting degree and accuracy, and can be used to predict the closing time of the thermal screen.

3.3. Performance Evaluation

Figure 10a shows the control method of the two-factor control method when the thermal screen turns on. The greenhouse control system obtains indoor and outdoor temperatures and calculates the temperature difference between the inside and outside air (X) through the detection equipment. Then, the internal and external temperature difference is brought into the function of opening to calculate the theoretical minimum solar radiation value of the opening thermal screen (Y₁). After that, Y₁ will be used to compared with the outdoor solar radiation value (Y₂). If the outside solar radiation (Y₂) is greater than the calculated value, then the thermal screen is opened. Otherwise, the process will be repeated after 10 min. Figure 10b shows the control process when the thermal screen is closed. The control method of closing is different from opening. When the outside solar radiation value (Y₂) is lower than the theoretical value (Y₁), the thermal screen is turned off.

To compare the two-factor control method with the traditional control method (opening at 8:30 and closing at 15:30), the greenhouse heat load under different control methods was calculated. In this paper, the traditional control method of thermal screens has no fixed method. Thus, the greenhouse heat load of conventional method in cloudy weather is considered identical to the new method, and the cooling point of the greenhouse is set at 50 °C, which is much higher than the natural air temperature of the greenhouse [25]. There is a vital temperature in crop cultivation, which is known as the biological zero-point, and the temperature is related to crop species. As one of the most widely grown crops, the biological zero-point of tomato is 8–10 °C. The crop stops this study of the minimum temperature, which the calculated heating load is set to 10 °C. The run period in this model is also set from 1 December to 31 January, which lasts for 62 days.

Figure 11 shows the greenhouse air temperature under different control methods in 5 days. The result shows that using the two-factor method can promote greenhouse temperature without consuming energy. The effect is most pronounced on the fifth day, because heat accumulation is a continuing process. The effect on temperature is most obvious when the thermal screen is opened and closed. In 5 days, there is a improve of 0.53 °C.

When compared with the traditional control method, the two-factor control method, which is based on the heat balance of the front roof, can significantly improve the thermal screen capacity of the greenhouse during the cold season, thus reducing the energy consumption of the greenhouse. According to the calculation, the heat load of the control method is 2.69 × 10⁶ MJ, which can save 2.12 × 10⁵ MJ in energy compared with the traditional control method of 2.75 × 10⁶ MJ. In this study, the energy savings rate is 7.2% (Figure 12). These results show that the automatic control of the opening and closing of the solar greenhouse thermal screen can not only effectively save labor costs, but also reduce the heat load of the greenhouse more effectively. However, this control method sometimes results in multiple openings and closings of the insulation tube in cloudy weather, and its effect on the mechanical structure of the thermal screen needs further evaluation.

4. Discussion

In this study, EnergyPlus, the outdoor temperature and physical parameters of greenhouse materials were used to build a dynamic prediction model of air temperature. The verification shows that the model has high accuracy, and the program can be used in the research of greenhouse energy saving.

The two-factor control method of thermal screens was studied under different weather conditions (two radiation controls and three temperature controls). The thermal balance of the front roof was considered when the thermal screen was opened. It was found that the influence of temperature gradient on the time of thermal screen control could reach 10 min, and the influence of light on the temperature in the greenhouse could reach 1 h. Therefore, the thermal insulation should be turned on to pay more attention to the outdoor solar radiation. The maximum heat storage of the heat accumulator was considered when the thermal screen was closed. It was found that the influence of temperature difference on the thermal screen was the same as the opening of the thermal screen when the outdoor radiation conditions were fixed. In this case, the closing time of the thermal screen was delayed by 0.5 h. This condition was caused by the low temperature difference between indoor and outdoor air (due to the low outdoor radiation throughout the day). The management method of thermal screen is determined by the outdoor environment temperature and solar radiation in the greenhouse, and the combined influence of two factors should be considered.

In this study, a two-factor thermal screen control method is established, which considers both temperature differences between indoor and outdoor and outdoor solar radiation, and it allows for operators to control greenhouses more flexibly. The calculation and the result show that the two-factor control method can save a part of the energy in a standard meteorological year. The advantage of this method is that it can effectively improve the thermal performance of the greenhouse with little cost and save the energy of the greenhouse in the cold season without extra investment. It is worth noting that the thermal screen should be closed earlier, in rain and snow weather, to prevent mechanical damage.

The objective of the two-factor control strategy is to obtain the optimal thermal performance of the greenhouse and does not consider the effect of light duration of the plants. Therefore, for some long-day plants, the duration of illumination can be properly extended or the plants should be given artificial light.

5. Conclusions

In this paper, numerical simulations and experimental studies were employed to examine the opening and closing of thermal screens in solar greenhouses in China. A CSG model was created using EnergyPlus software, and the greenhouse air temperature of the greenhouse was simulated. The model was calibrated with experimental data, which were used in predicting greenhouse air temperatures.

The opening and closing times of the thermal screen under different weather conditions were calculated, and the optimal control strategy was explored. The reasonable opening and closing times of greenhouse heat preservation in December and January were calculated by using typical meteorological years. The results show that the temperature has little effect on the opening and closing times of the thermal screen, but the solar radiation has a great effect.

The differences and outdoor solar radiation were correlated to establish a control function, and the control function of the thermal screen was established. The effectiveness of the traditional control method was compared with that of the two-factor control method. The results showed that the two-factor control method can save 7.2% of energy consumption in winter. These models can provide a reference for the opening and closing of greenhouse thermal screens in the cold regions of China.