Comparison of the feedback linearization plus LQR controller and the PID controller for greenhouse indoor climate

Greenhouse environmental control systems can improve the growth and quality of the plants within greenhouses by keeping a constant environment. Greenhouse climate is a multi-input multi-output system that is significantly affected by climate factors like temperature, relative humidity, and carbon dioxide levels. Due to the nonlinearity and existence of coupling among climate factors, the designed controller should provide good control performances. This study proposed both the feedback linearization plus linear quadratic regulator (LQR) controller and the proportional-integral-derivative (PID) controller for indoor air temperature and humidity control of a greenhouse system. The nonlinear greenhouse model was transformed into its equivalent linear form using input-output feedback linearization. Then, a proportional-integral type LQR controller was designed for the linear form to achieve the overall nonlinear feedback control law. In addition, the practical PID controller was designed and its gains were tuned using a genetic algorithm by considering the integral of absolute error and control deviation, and the integral of squared error and control deviation. A set of simulation works done on the nonlinear model illustrates the effectiveness of the two control methodologies. Two control methods, feedback linearization plus LQR and PID, demonstrated effective performance in both setpoint tracking and disturbance rejection. The feedback linearization plus LQR controller exhibited superior disturbance rejection capabilities, characterized by reduced perturbation peaks and faster recovery times. Conversely, the PID controller demonstrated superior setpoint tracking performance with minimal overshoot.