Research On Inverse Model Decoupling Control System Of Temperature And Humidity In Fruit And Vegetable Greenhous

China is a major player in fruit and vegetable production,with greenhouse cultivation being the primary method employed.Precisely controlling the environmental factors in greenhouses is vital for increasing crop yields,improving fruit quality,and mitigating plant diseases and mortality.Research indicates a strong coupling between temperature and humidity,coupled with noticeable lag in their changes within the greenhouse environment.Consequently,independently controlling temperature and humidity presents challenges.To enhance control precision and minimize their mutual interactions,a reverse model decoupling control strategy has been proposed.This strategy aims to achieve linearized decoupling control of temperature and humidity in fruit and vegetable greenhouses,thereby improving overall control accuracy.This research utilizes machine learning and BP neural networks in conjunction with the principles of reverse model decoupling control to identify the inverse model of the temperature and humidity environment.The outcome is a more precise inverse model.A comparative analysis is performed between traditional fuzzy PID and variable domain fuzzy PID.Subsequently,a reverse composite control system is developed using the inverse model,enabling decoupling control of temperature and humidity within the original greenhouse environment system.This methodology enhances control accuracy and reduces response time.Through experimental validation of the designed control strategy for real-world fruit and vegetable greenhouse environments,the effectiveness of decoupling control for temperature and humidity is demonstrated.The primary research content encompasses the following aspects:(1)Environmental data related to the research subject,including temperature,humidity,heating water pipe temperature,dehumidification fan air velocity,and light intensity,are collected.These data are integrated with the data from the daylight greenhouse and processed accordingly.(2)The original system is subjected to inverse identification using BP neural networks and least squares support vector machines.A comparative analysis is performed between these two methods of inverse system identification,and the superior approach is selected.The resulting inverse model from the identification process is then connected in series with the original system,creating a pseudo-linear composite system that serves as the foundation for decoupling control.(3)A comprehensive study is conducted on the control aspect of the system.A comparison is made between fuzzy PID control and variable domain theory control.The performance of the PID controller is thoroughly analyzed based on simulation results.The inverse identification model and the PID control model are combined to create a composite controller.(4)Following the research methodology,a greenhouse environment test platform is constructed,and experiments are carried out in an actual greenhouse setting.By analyzing the outcomes of these experiments,the effectiveness of the data is verified.The greenhouse environment is effectively controlled using a combination of the heat accumulation control strategy and the reverse model decoupling control method specifically designed for fruit and vegetable greenhouses.This control approach ensures that the temperature within the greenhouse remains within the desired range of 16℃ to 28℃,while the humidity is maintained between 60%RH and 90%RH.Notably,the temperature control achieves a significantly faster response time compared to traditional control methods,reducing the time required by approximately 1000 seconds.Similarly,the humidity control demonstrates improved speed,with a response time reduced by around 700 seconds when compared to traditional approaches.Moreover,the temperature control exhibits a remarkable accuracy,with an error rate of approximately 1.2%,while the humidity control maintains a commendable error rate of around1.5%.