Analysis And Simulation Of Microclimate Of Single-span Plastic Greenhouse For South China In Autumn And Winter

The internal environmental conditions including solar radiation, temperature, relative humidity and CO2 concentration are important to crops growth because the greenhouse looks like a closed agro-ecosystem.In this paper, the spatial and temporal characteristics of interior microclimate of single-span plastic greenhouse for South China in autumn and winter were analyzed. Taking full account of the thermal effect and dynamic effect of plants to indoor environment, the numerical simulations of environmental factors for plastic greenhouses with diverse covering layers and management methods were carried out using statistical models, physical models and computational fluid dynamics models. The main research contents and conclusions are as follows:1. The spatial and temporal characteristics of internal environmental conditions including solar radiation, air temperature, relative humidity, soil temperature and CO2 concentration of greenhouses with two cover types indicated that the diurnal variations of indoor meteorological factors except CO2 concentration were similar to that of outdoor meteorological factors in same height. Solar radiation, air temperature and soil temperature increased after sunrise, peaked in noon, and then decreased. While relative humidity showed the opposite pattern that first decreased after sunrise, and then gradually increased. Outdoor CO2 concentration presented only small fluctuations during a day. However, indoor CO2 concentration showed a significant diurnal variation that declined after sunrise and gradually rose until from 15:30 pm with the solar radiation and crop photosynthesis becoming strong to weak. The differences between indoor factors and corresponding outdoor factors also fluctuated during a day and the variation patterns were similar to that of the own factors. It was found that indoor air temperature and relative humidity differed from north to south. The air temperature of south part was greater than that of the north part, while the relative humidity of south part was smaller than that of the north part. The difference between east part and west part was not obvious. Double-layer covering greenhouse can offer higher air temperature and soil temperature compared to single-layer covering greenhouse, and was beneficial to crops growth in winter. But the solar radiation was smaller and the relative humidity was greater in double-layer covering greenhouse than in single-layer covering greenhouse. Therefore, it is necessary to take reasonable adjustment measures for its indoor environmental characteristics when using multi-layer covering greenhouse.2. The average values and the extreme values of indoor air temperature and relative humidity were both significantly related to the outdoor meteorological elements at 5% significance level. So it is reasonable and feasible to use statistical methods for simulating and predicting the indoor environmental factors based on the outdoor weather elements. Making full use of the historical indoor temperature and relative humidity data, multiple-regression method combined with autoregressive integrated moving average model (ARIMA model) was used to simulate and predict the indoor temperature and relative humidity, especially the extreme values of these factors. The accuracy of new method was higher than the traditional multiple-regression method and the BP neural network by analyzing the effect of three time scales including ten minute, one hour, and one day. This new method avoiding the short come of poor stability of the BP neural network method can be extended in the actual production weather forecast.3. The plastic greenhouse was divided into covering layer, indoor air, plant canopy and soil layer. Based on the mass and energy balance between indoor and outdoor environments, the physical models were established to estimate the air temperature and relative humidity for single-layer covering greenhouse and double-layer covering greenhouse respectively and were verified by measured meteorological data. The mean relative errors between simulated values and measured values of air temperature, relative humidity and soil temperature were less than 7.0%,8.0% and 5.0% respectively, indicating that the physical models exhibited high accuracy. The highest accuracy was found for overcast sky, followed by cloudy day.4. Based on the indoor and outdoor observed data, the simulation and verification of air temperature and relative humidity were carried out under ventilation and non-ventilation conditions by the use of computational fluid dynamics (CFD) technology. The mean relative errors and the root mean square error between simulated values and measured values of air temperature were 4.47% and 1.34℃under non-ventilation conditions. The mean relative errors and the root mean square error between simulated values and measured values of air temperature were 4.68% and 1.54℃, and were 5.22% and 3.54% for relative humidity under ventilation conditions. The spatial distributions of air temperature and relative humidity simulated by CFD models were consistent with the observed distributions, showing that this method is feasible and reasonable in the greenhouse environment simulation. The results fully demonstrating the differences of temperature and relative humidity between diverse vertical heights and diverse directions provided a data base for local control and refine management of greenhouse environment.5. The ventilation effects under different ventilation time, different ventilation velocities and different ventilation sizes were discussed based on the computational fluid dynamics (CFD) technology. With the increase of ventilation time, ventilation velocities and ventilation sizes, the cooling effect of ventilation played a role gradually from the north part to the south part, from lower part to higher part in the plastic greenhouse. However, the difference between decreasing magnitudes of temperature varied with the constant gradient of the three parameters, indicating that the cooling effect caused by the variation of these three parameters will be gradually weakened when the mixture of indoor and outdoor air was steady. Actual natural ventilation condition was more complex than the CFD experiment. Although the CFD simulation can not completely reflect the reality, the results were consistent with the physical laws and theoretical expectations and able to provide more refined data in three-dimensional space. It can be used to offer a theoretical basis for the control and optimization of greenhouse environment.