Desarrollo y evaluación de un control neuro-difuso tipo ANFIS frente a un control PID convencional aplicado al péndulo invertido
This is a research and development project aimed to develop a neuro-fuzzy controller, general concepts, characteristics and its classification grouped under ANFIS type neuro-fuzzy control will be used to design an algorithm to control this project. The programming of the controller and the developme...
| Autor Principal: | Jara Chico, María José |
|---|---|
| Otros Autores: | Rocha Proaño, Diego Raúl |
| Formato: | bachelorThesis |
| Idioma: | spa |
| Publicado: |
2017
|
| Materias: | |
| Acceso en línea: |
http://dspace.ups.edu.ec/handle/123456789/13353 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: |
This is a research and development project aimed to develop a neuro-fuzzy controller, general concepts, characteristics and its classification grouped under ANFIS type neuro-fuzzy control will be used to design an algorithm to control this project. The programming of the controller and the development of a graphical interface for displaying parameters in both steady state and transitory regime will be accomplished using MATLAB software. Once the ANFIS type neuro-fuzzy controller is developed, a software level a study will be performed on the plant on which the controller is going to be operated to validate the performance of the controller. It will cover the operating characteristics of the component present in the module names “Digital Pendulum” Furthermore, the mathematical modeling of the plant will be established and the control variables will be identified. In order to implement the ANFIS type neuro-fuzzy controller a plant that is already located at the Laboratory of Control Theory of the Polytechnic Salesian University, South Campus. When the ANFIS type neuro-fuzzy controller is implemented, data will be collected in order to make a comparative analysis between the current PID controller that is already installed at the plant against the ANFIS type neuro-fuzzy controller developed for this project and with the analysis of the results, a list of advantages and disadvantages for each controller will be established. For each of the following design parameters: maximum peaks, settling time, steady-state error, stability, etc. |
|---|