| In winter,the natural environment in the northern region is not suitable for planting crops,but there is still a large demand for fresh fruits and vegetables in the market.The solar greenhouse can provide suitable growth environment for fruits and vegetables,which can not only meet the consumption demand,but also solve the problem of long-distance transportation and preservation,and stabilize the market price of fruits and vegetables in winter in northern China.The indoor environment of greenhouse is greatly affected by the external environment,and the indoor microclimate environment has an important impact on crop growth.Therefore,understanding the spatial distribution of indoor temperature and humidity,and then predicting and controlling indoor temperature and humidity,can provide suitable growth environment for crops,ensure the quality of fruits and vegetables,and increase crop yield and farmers’ income.This thesis takes the double-layer film greenhouse as the research object,analyzes the collected data inside and outside the greenhouse,and studies the distribution law of temperature and humidity inside the solar greenhouse.The BP neural network prediction model was established by LM algorithm,and the temperature and humidity in the greenhouse were predicted by the whole-day method and the segmental method respectively.The simulation model of traditional PID control and fuzzy PID control was established in Simulink to compare the control effect of temperature and humidity.The main conclusions are as follows:(1)The temperature and humidity in the greenhouse changes steadily at night.In the north-south direction,the farther away from the front foot of the greenhouse,the higher the temperature and the lower the humidity.In the east-west direction,the farther away from the west wall of the greenhouse,the higher the temperature and lower the humidity.With the increase of the height from the ground inside the greenhouse,the temperature is higher,and the humidity is lower.During the day,with the increase of outdoor temperature and solar radiation,the indoor temperature rises and the humidity decreases.In the north-south direction,the closer the indoor soil is to the front foot of the greenhouse the lower the temperature and the higher the humidity.(2)The compliance and root mean square error of the predicted indoor air temperature and humidity for the whole day were 0.9842,0.9846 and 2.3655,4.5092 respectively.The compliance and root mean square error of the predicted indoor air temperature and humidity for the nighttime were 0.9968,0.9579 and 0.3655,0.2759 respectively.The compliance and root mean square errors for daytime indoor air temperature and humidity were 0.9979,0.9999 and 0.9965,1.2272 respectively.A comparison of the two prediction methods revealed that the segmented prediction method errors were smaller.(3)Compared with the traditional PID control and fuzzy PID control method,the temperature and humidity control of solar greenhouse is simulated.The control performance of the fuzzy PID control simulation model is better according to the analysis of performance index such as rise time,overshoot and adjustment time.The fuzzy PID control model has a temperature and humidity rise time of 302.7ms,204 ms,an overshoot of 4.00%,16.28% and a regulation time of 955 ms,830ms.Compared with the traditional PID control model,the temperature and humidity rise times are 85.3ms and 13.5ms faster,the overshoot is 2.07%and 3.98% smaller,and the adjustment times are 624 ms and 411 ms faster. |