Greenhouse Temperature Based On PIDNN Control And Research

Temperature is an extremely important parameter to maintain crop growth.Greenhouse is a key facility in modern agriculture that can provide suitable growth temperature environment for crops in different seasons and different climates.The traditional PID control method is used to control the temperature of the greenhouse.This kind of method has a low level of automation and intelligence.It relies too much on manual experience and is inefficient.control.Therefore,this paper proposes a control algorithm(GA-MPIDNN)that combines genetic algorithm and improved PID neural network to achieve high-efficiency,high-precision,and high-stability control of plastic greenhouse temperature.The main research work of this paper is summarized as follows:(1)The form of the transfer function of the dynamic mathematical model of the greenhouse is deduced according to the law of heat balance in nature,and then by injecting a step signal into the plastic greenhouse,and using a two-point fitting algorithm to perform mathematical operations according to the step response curve,the model is obtained.The actual values of the parameters are finally obtained to obtain the specific dynamic mathematical model of the greenhouse under this environment.(2)For the greenhouse environment with large inertia and large time delay,this study uses the obtained mathematical model of the greenhouse temperature control system to try to introduce a neural network with self-learning characteristics into the PID controller(PIDNN),and combine the genetic algorithm to initialize the most The optimal weight value is added to improve the back-propagation operation of the BP network by adding a momentum term to form a GA-MPIDNN controller with good convergence and self-adaptive characteristics,which is then regulated in the simulation environment of greenhouse temperature.The results show that: based on the ZN method and the genetic algorithm,the GA-MPIDNN algorithm effectively reduces the overshoot of the control process,shortens the control time,and at the same time improves the stability of the system,and improves the dynamic performance of the control system in many aspects.(3)The GA-MPIDNN algorithm,which showed the optimal control performance in the simulation experiment,was introduced into the real environment,and an intelligent temperature control system for small greenhouses based on GA-MPIDNN was built.The upper computer of the system is built by using the serial communication between Python and the single-chip microcomputer,and the lower computer of the system is built by combining the temperature information transmission between the sensor and the single-chip microcomputer,and the control signal transmission between the single-chip microcomputer and the actuator.The target temperature for regulation is set,the temperature regulation of the system is carried out in the real environment and the test results are obtained.The test results show that the system has successfully achieved intelligent temperature control and achieved a relatively ideal regulation effect.