Estudio para el uso de inteligencia artificial en el control de una torre de destilación extractiva
Tradeoff between precision and resolution time in the mathematical models used in the nonlinear predictive control has been one of the biggest challenges of this technique. Additionally, models that not use the real system data may not represent the system at all. Intelligent agents are AIs systems...
- Autores:
-
Bernal Acosta, Jairo Iván
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2018
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/61860
- Acceso en línea:
- http://hdl.handle.net/1992/61860
- Palabra clave:
- Alcohol
Destilación extractiva
Inteligencia artificial
Redes neurales (Computadores)
- Rights
- openAccess
- License
- http://creativecommons.org/licenses/by-nc-nd/4.0/
Summary: | Tradeoff between precision and resolution time in the mathematical models used in the nonlinear predictive control has been one of the biggest challenges of this technique. Additionally, models that not use the real system data may not represent the system at all. Intelligent agents are AIs systems capable of learning from real data, identify patterns and predict variables in time. That is, they are capable of mixing real data with mathematical models. In the present article, an intelligent agent was implemented to replace the complex mathematical model in an extractive distillation tower to study the possibility to use it as an auxiliary control method. A neuronal net was trained to predict the mathematical model with a precision of 98% from data obtained in a simulation of the distillation of azeotropic ethanol. |
---|