two main fields (Duarte-Galvan et al. 2012). The first one is usually called
conventional control which consists on climate controllers which try to control the
greenhouse microclimate just by reducing the deviation between setpoint of the
interest variables and measured values to zero. As examples of conventional
controllers are ON/OFF, PID, other classical controllers and also artificial intelligent (AI) paradigms such as artificial neural networks (ANNs), fuzzy logic
systems (FLS), genetic algorithms (GAs), among others. The other field is optimal
control, in which factors such as greenhouse dynamic behavior, actuator capabilities, water and energy consumption and meanly the crop response are taken
into account. Expert systems and model predictive control (MPC) are widely
accepted for optimal control purposes. However, aforementioned AI-based techniques can be also considered like optimal production controllers when they
reached objectives such us optimal crop growth, reduction of the associate costs,
reduction of residues and the improvement of energy and water use efficiency
(Ramírez-Arias et al. 2012).
Conventional controllers were widely used since computational tools were
introduced in protected agriculture until the end of twentieth century. Nevertheless, the increase in power computational capabilities with cost reductions in the
next decade allowed the application of more complex algorithms which deals with
the optimal control scheme. In the rest of this chapter, the review of protected
agriculture techniques will be focus on optimal control taking into consideration
both modeling and controllers.
13.11.3 Optimal Control
The objective of modern greenhouse industry is a sustainable crop production
system by reducing water and energy consumption and biocide use while maintaining a high crop quality and yield. In addition to new materials and advance
building methodologies, this can be achieved by modifying the microclimate
control strategy. However, the optimal control scheme is only slowly accepted in
practice due to the lack of reliable crop development models for the wide variety
of crops. Also, experimental probes that show advantages and clear assessment of
the risks involved and the theoretical limitation is required. In order to effectively
validate the performance of proposed optimal control algorithms, greenhouse
models which include parameters related with crop response are required.
13.11.4 Modeling
A good dynamic model of the process is essential in order to achieve a good
performance of the controller. Two different methods for computing greenhouse
models can be found in the literature. One is based in terms of the physical laws
13 Instrumentation and Control to Improve the Crop Yield
389
conventional control which consists on climate controllers which try to control the
greenhouse microclimate just by reducing the deviation between setpoint of the
interest variables and measured values to zero. As examples of conventional
controllers are ON/OFF, PID, other classical controllers and also artificial intelligent (AI) paradigms such as artificial neural networks (ANNs), fuzzy logic
systems (FLS), genetic algorithms (GAs), among others. The other field is optimal
control, in which factors such as greenhouse dynamic behavior, actuator capabilities, water and energy consumption and meanly the crop response are taken
into account. Expert systems and model predictive control (MPC) are widely
accepted for optimal control purposes. However, aforementioned AI-based techniques can be also considered like optimal production controllers when they
reached objectives such us optimal crop growth, reduction of the associate costs,
reduction of residues and the improvement of energy and water use efficiency
(Ramírez-Arias et al. 2012).
Conventional controllers were widely used since computational tools were
introduced in protected agriculture until the end of twentieth century. Nevertheless, the increase in power computational capabilities with cost reductions in the
next decade allowed the application of more complex algorithms which deals with
the optimal control scheme. In the rest of this chapter, the review of protected
agriculture techniques will be focus on optimal control taking into consideration
both modeling and controllers.
13.11.3 Optimal Control
The objective of modern greenhouse industry is a sustainable crop production
system by reducing water and energy consumption and biocide use while maintaining a high crop quality and yield. In addition to new materials and advance
building methodologies, this can be achieved by modifying the microclimate
control strategy. However, the optimal control scheme is only slowly accepted in
practice due to the lack of reliable crop development models for the wide variety
of crops. Also, experimental probes that show advantages and clear assessment of
the risks involved and the theoretical limitation is required. In order to effectively
validate the performance of proposed optimal control algorithms, greenhouse
models which include parameters related with crop response are required.
13.11.4 Modeling
A good dynamic model of the process is essential in order to achieve a good
performance of the controller. Two different methods for computing greenhouse
models can be found in the literature. One is based in terms of the physical laws
13 Instrumentation and Control to Improve the Crop Yield
389