基于神经网络PID的番茄室温环境控制

Control of Tomato Room Temperature Environment Based on Neural Network PID

番茄在温室环境中种植时,温度、湿度、CO2浓度是影响番茄品质的重要因素。番茄温室环境通常是一个非线性、时变性、滞后性复杂系统,采用传统PID对温度、湿度、CO2浓度进行控制效果并不理想。为提高番茄温室环境控制效果,设计一种基于神经网络PID的自适应控制方法。介绍番茄温室控制方案,结合控制方案设计控制系统硬件。为提高温室系统自适应能力,将神经网络自学习算法与PID算法相结合,实现PID参数的在线自适应调整。仿真结果表明,该控制方法与传统PID控制方法相比,大幅降低系统收敛时间,控制精度得到大幅提高。

When tomato was planted in greenhouse environment, temperature, humidity and CO2 concentration were important factors affecting tomato quality. Tomato greenhouse environment was usually a non-linear, time-varying and hysteretic complex system, and the traditional PID control of temperature, humidity and CO2 concentration was not ideal. In order to improve the environmental control effect of tomato greenhouse, an adaptive control method based on neural network PID was designed. The control scheme of tomato greenhouse was introduced, and the hardware of the control system was designed combined with the control scheme. In order to improve the adaptive ability of greenhouse system, the self-learning algorithm of neural network and the PID algorithm were combined to realize the on-line adaptive adjustment of the PID parameters. The simulation results showed that compared with the traditional PID control method, the control method greatly reduced the system convergence time and improved the control accuracy.