14.3.3 Evolutionary and Bio-Inspired Control
Evolutionary algorithms (Genetic algorithms, Evolution strategies, Evolutionary
programming, and Differential evolution) are heuristic global search methods
inspired by natural evolution and genetics (Eiben and Smith 2003). These algorithms
have been applied recently to optimize and control the greenhouse environment.
Guzmán-Cruz et al. (2009) used Genetic Algorithms (GAs), Evolutionary Strategies
(ES), and Evolutionary Programming (EP) to calibrate parameters of a climatic
model that describes the behavior of air temperature and relative humidity inside a
greenhouse. Other authors have used genetic algorithms to carry out the parameter
estimation for greenhouse crop growth model (Dai et al. 2009). Herrero et al. (2007)
performed a nonlinear robust identification of a greenhouse model using a multiobjective genetic algorithm (MOGA). A MOGA to estimate and optimize the
parameters of a pipe heating system inside a greenhouse was presented by Braga
(2006). This algorithm is focused on heat radiation transfer. A robust identification
of a nonlinear greenhouse model has been carried out using an evolutionary algorithm to solve a complex multimodal optimization problem (Herrero et al. 2008). An
optimal control problem of the greenhouse climate aiming maximization of the
profit was stated and solved using two evolutionary algorithms (Pohlheim and
Heibner 1997). Both a multi-strategy competition and an evolution strategy evolutionary algorithm were evaluated (Pohlheim and Heibner 1999). A classical
genetic algorithm was used as the optimizer in an MPC problem to optimize energy
and water consumption (Blasco et al. 2007). A differential evolution algorithm was
used to solve an optimal control problem of lettuce cultivation in a greenhouse
(López-Cruz et al. 2003a, b). Evolutionary algorithms were applied recently to solve
multiobjective optimization problems of the greenhouse environment (Hu et al.
2011b). Bio-inspired algorithms (Particle Swarm, Ant colony, and Artificial Bee
colony optimization) are also heuristic global search methods inspired by the
complex behavior of biological systems (Yang 2008) that can be used in the control
of the greenhouse system. A particle swarm optimization algorithm was used in the
control of greenhouse temperature using MPC approach (Coelho et al. 2005).
References
Abdhel HI (2010) New applications and developments of fuzzy systems. PhD Thesis, Korea
University, Seoul, p 95
Alghannan A (2000) Using proportional integral derivative and fuzzy logic with optimization for
greenhouse climate control. PhD Dissertation, University of Idaho, USA, p 136
Astrom KJ, Wittenmark B (1994) Adaptive control. Addison-Wesley, Englewood Cliffs
Arvanitis KG, Paraskevopoulus PN, Vernardos AA (2000) Multirate adaptive temperature control
of greenhouses. Comput Electron Agric 26:303–320
Athans M, Falb PL (2007) Optimal control. An introduction to the theory and its applications.
Dover Publications, Inc, New York
14 Control Strategies of Greenhouse Climate
415
Evolutionary algorithms (Genetic algorithms, Evolution strategies, Evolutionary
programming, and Differential evolution) are heuristic global search methods
inspired by natural evolution and genetics (Eiben and Smith 2003). These algorithms
have been applied recently to optimize and control the greenhouse environment.
Guzmán-Cruz et al. (2009) used Genetic Algorithms (GAs), Evolutionary Strategies
(ES), and Evolutionary Programming (EP) to calibrate parameters of a climatic
model that describes the behavior of air temperature and relative humidity inside a
greenhouse. Other authors have used genetic algorithms to carry out the parameter
estimation for greenhouse crop growth model (Dai et al. 2009). Herrero et al. (2007)
performed a nonlinear robust identification of a greenhouse model using a multiobjective genetic algorithm (MOGA). A MOGA to estimate and optimize the
parameters of a pipe heating system inside a greenhouse was presented by Braga
(2006). This algorithm is focused on heat radiation transfer. A robust identification
of a nonlinear greenhouse model has been carried out using an evolutionary algorithm to solve a complex multimodal optimization problem (Herrero et al. 2008). An
optimal control problem of the greenhouse climate aiming maximization of the
profit was stated and solved using two evolutionary algorithms (Pohlheim and
Heibner 1997). Both a multi-strategy competition and an evolution strategy evolutionary algorithm were evaluated (Pohlheim and Heibner 1999). A classical
genetic algorithm was used as the optimizer in an MPC problem to optimize energy
and water consumption (Blasco et al. 2007). A differential evolution algorithm was
used to solve an optimal control problem of lettuce cultivation in a greenhouse
(López-Cruz et al. 2003a, b). Evolutionary algorithms were applied recently to solve
multiobjective optimization problems of the greenhouse environment (Hu et al.
2011b). Bio-inspired algorithms (Particle Swarm, Ant colony, and Artificial Bee
colony optimization) are also heuristic global search methods inspired by the
complex behavior of biological systems (Yang 2008) that can be used in the control
of the greenhouse system. A particle swarm optimization algorithm was used in the
control of greenhouse temperature using MPC approach (Coelho et al. 2005).
References
Abdhel HI (2010) New applications and developments of fuzzy systems. PhD Thesis, Korea
University, Seoul, p 95
Alghannan A (2000) Using proportional integral derivative and fuzzy logic with optimization for
greenhouse climate control. PhD Dissertation, University of Idaho, USA, p 136
Astrom KJ, Wittenmark B (1994) Adaptive control. Addison-Wesley, Englewood Cliffs
Arvanitis KG, Paraskevopoulus PN, Vernardos AA (2000) Multirate adaptive temperature control
of greenhouses. Comput Electron Agric 26:303–320
Athans M, Falb PL (2007) Optimal control. An introduction to the theory and its applications.
Dover Publications, Inc, New York
14 Control Strategies of Greenhouse Climate
415