Control solutions for greenhouse climate systems

Abstract

The PhD thesis develops solutions for greenhouse climate control systems (temperature and humidity control). This is an actual subject included in the more larger domain of application of control system in agriculture. The climate control system using a nonlinear-coupled model, with important influence of external disturbances, is implemented employing linearization and decoupling module and mainly PID controllers tuned by different methods. The feedback-feedforward linearization and decoupling procedure applied to the greenhouse climate model requires the estimation of the output vector without delay, which is estimated by employing a state observer proposed by the author. An integrator plus dead time (IPDT) behavior is obtained for the greenhouse climate process with linearization and decoupling. There are developed greenhouse climate control solutions, using the equivalent IPDT decoupled, linearized model, by employing PID controllers tuned by: five conventional methods, three two degree of freedom (2DoF) methods, and genetic algorithms with proposed specific cost functions. The control system performances employing the previously mentioned controllers are analyzed and compared using the simulation results. A modified Smith predictor using a proposed state observer is also developed and tested by simulation. Finally, the thesis focuses on the development and the implementation of a Hardware-in-the-Loop (HIL) - SCADA platform for greenhouse climate control using NI LabVIEW and Siemens industrial automation software package and Siemens hardware technology. SCADA specific modules, i.e., two Human Machine Interfaces (HMI), and a generic GSM Ethernet Telematics System have been also developed.