Research And Design On Greenhouse Environment Decoupling Control Based On Neural Network

In recent years,more and more advanced control methods have been introduced into agricultural production and planting,which has led to the rapid development of environmental control technologies in greenhouses.As the control system of environmental variables in greenhouses is a relatively complex and nonlinear dynamic system,the controlled objects in the greenhouse interact with each other and the environmental variables interact with each other.At the same time,the growth of crops also affects the environmental changes in greenhouses.Make the greenhouse environmental variable control into a MIMO system with much factor.In this way,it is difficult to achieve ideal control effects when using conventional control methods to control environmental variables in greenhouses.In this paper,the various factors are controlled by the environmental variables inside the greenhouse.The coupled control system is studied and a decoupling control method based on BP neural network and PID control technology is applied to the greenhouse environment variable control system.This paper first studies and analyzes several kinds of multivariable decoupling control methods widely used in the control field,and summarizes their advantages and disadvantages.Then using the powerful learning ability and adaptability of BP neural network combined with PID control algorithm,a multivariable PID neuron decoupling controller based on BP neural network was built to control the characteristics of greenhouse environmental variables.The model of environment variable control system for greenhouses was established by using step response curve method.The PID neuron decoupling controller was modeled and simulated by Matlab.By contrast,it was found that the decoupling of PID neurons was used to control the environmental variables of greenhouses.The device has good decoupling and control effects.Finally,using embedded control technology to develop a greenhouse environment variable control and management system,which integrates environmental data collection and storage,greenhouse environmental control,and human-computerinteraction control.And set up a greenhouse laboratory platform.Through the intranet penetration technology,the communication between the laboratory platform and the host computer is achieved through the TCP/IP protocol.Using Delphi software,a remote control system for environmental variables in greenhouses was designed.The decentralized control of environmental variables in greenhouses was carried out through an experimental platform.The test results show that the control effect of the decoupled controller using PID neurons in the actual control meets the requirements of the actual greenhouse greenhouse environmental variable control.