Modelación y control de un invernadero experimental.

Agricultural systems such as greenhouses are difficult to control with classical regulators as a consequence of their great complexity and their non-linear dynamic behaviour. This project has been carried out with the purpose of modeling and controlling an experimental greenhouse, applying intelligent control such as neural networks, which allows controlling the environmental variables of temperature and humidity at the right time and thus the crop does not lose its properties, in addition, the loss of the crop will be avoided. One of the main objectives of this work is to solve the problems that arise in automated greenhouses. The transfer functions of the data obtained from temperature and humidity are used to obtain the appropriate models. The control that is implemented is an advanced control of neural networks NARMA-L2, which allows to control the plant in an adequate way. The results indicate a good and stable performance of the proposed controller. The security provided by the control system is appropriate since the data obtained are those desired in the simulation, the relationship of temperature and humidity can be simulated for different possible scenarios, guaranteeing the proper functioning of the control. The errors in stable state tend to zero, having a temperature steady state error of 0.03 and humidity with an error of 0.0036, allowing to improve the performance of the greenhouse.

• Main Author: Molina Fernández, María Fernanda (author)
• Format: masterThesis
• Language: spa
• Published: 2022
• Subjects: CONTROL DE INVERNADEROS; REDES NEURONALES; AUTOMATIZACIÓN DE INVERNADEROS; ELECTRICIDAD
• Online Access: http://repositorio.utc.edu.ec/handle/27000/9560