Study On Control Method Of Plant Growth Cabinet

In recent years,with the improvement of people’s living standard and the continuous development of social economy,the contradiction between the demand for green pollution-free vegetables and the continuous reduction of agricultural arable land area is increasingly prominent.Plant factory is an integrated innovation product of modern science and technology in the field of agriculture.It can achieve multiple goals of high quality,ecology,high efficiency,high yield,safety and ecology,which is the direction of future agricultural development.In the plant factory,the temperature is a key factor in deciding growth of plants,but the temperature is a complex object with time-varying,nonlinear,large delay and unidirectional rising.The control effect of the conventional PID controller is often not good.BP neural network has strong self-learning and adaptive ability.The combination of BP neural network intelligent algorithm and conventional PID control has a better effect on temperature control in plant factories,and it can effectively improve the unstable control effect caused by external uncertainties.This paper mainly researchs the control method of temperature in plant factory.The conventional PID has large overshoot,control parameter tuning difficult,year long adjustment time and poor anti-interference ability in the process of temperature control,so the paper proposes combination of BP neural network and conventional PID,which is used to find the PID control parameters that can make the control system achieve the best performance based on the self-learning characteristics of the control system performance,and the control parameters are given to the conventional PID controller.In order to verify the difference in control performance of the conventional PID controller and the BP neural network PID controller designed in this paper,the mathematical model of the temperature in plant factory is established,and matlab software is used to simulate and analyze the mathematical model of temperature under the control of the PID controller based on BP neural network and the conventional PID controller.The simulation results show that compared with the conventional PID controller,the BP neural network PID controller designed in this paper has smaller overshoot and shorter regulation time in the control process,and it can adjust the external disturbances in time,thus improving the control performance.Finally,in this paper,a small plant growth cabinet is designed with STM32MCU as the control core.In the plant growth cabinet,various sensors measure the internal environment and transmit them to the single chip microcomputer.By comparing and analyzing the setting values of measured values and touch-screen,a single chip microcomputer controls the actuators of compressors,heating rods,CO₂ generators,solenoid valves,humidifiers,liquid temperature heating refrigerators,fans and pumps,therefore the plant cabinet can control the air temperature and humidity,the temperature of the nutrient solution,the wind speed,the intensity of light and the concentration of CO_2,which provides a suitable growth environment for plant in the plant cabinet.