| Greenhouse control is the effective production tools in the process of facilities agricultural cultivation, which has become an important research topic at home and abroad. Greenhouse environment, associated with the interaction among the various environmental factors which are also affected by crop growth, is a dynamic and extremely complex system. Therefore, how to achieve the coordination control of greenhouse among the various environmental factors, while reducing the operating costs of equipment, will be the main problem the greenhouse control needs to solve.In this paper, the modeling and control algorithms greenhouse environmental are deeply studied on the basis of the relevant literature. According to the law of heat balance and mass balance, the mathematical model of greenhouse temperature and humidity can be built. In the winter, the conventional PID, pure fuzzy, fuzzy PID controllers are designed to control the greenhouse temperature combined with the feature of temperature model, not consider the coupling effect between temperature and humidity. The simulation results show that fuzzy PID controllers can achieve the best control effect. In the summer, the feedback feed-forward linearization and decoupling and PID neural network are adopted to design the decoupling controller, with the coupled and nonlinear characteristics of greenhouse temperature and humidity model. The MATLAB simulation experiment results display that the designed controllers have good performance, can accurately track the target curve of the temperature and humidity. However, the feed-feedback control algorithm is mathematically precise control, which greatly depends on the accuracy of the greenhouse model. The neural network does not relay on the mathematical model with strong robustness and fault tolerance, has a stronger adaptability to the properties of the greenhouse complex mechanism. |
