Entrenamiento de una red neuronal artificial para el control de voltaje y reducción de pérdidas en la red de subtransmisión de 69kV de la EERSA
In the present work, the training and evaluation of neural networks is carried out for the improvement of voltage profiles and loss reduction through capacitive compensation, applying it to the 69kV system of Empresa Eléctrica Riobamba S.A. The training process begins with the development of a optim...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2023 |
| País: | Ecuador |
| Institución: | Universidad Técnica de Cotopaxi |
| Repositorio: | Repositorio Universidad Técnica de Cotopaxi |
| Idioma: | español |
| OAI Identifier: | oai:oai:repositorio.utc.edu.ec:27000:27000/10015 |
| Acceso en línea: | http://repositorio.utc.edu.ec/handle/27000/10015 |
| Access Level: | acceso abierto |
| Palabra clave: | COMPENSACIÓN OPTIMIZACIÓN RED NEURONAL PROBABILIDAD ELECTRICIDAD |
| Sumario: | In the present work, the training and evaluation of neural networks is carried out for the improvement of voltage profiles and loss reduction through capacitive compensation, applying it to the 69kV system of Empresa Eléctrica Riobamba S.A. The training process begins with the development of a optimization methodology that will determine the best values of reagents to be injected at each instant and in each of the bars, the different operating events that can occur in the system will be determined probabilistically and the optimization methodology will be applied to generate an output response to each event evaluated. Finally, the training process of the RNAs will take previously determined events and responses as input and output parameters, the trained networks will be evaluated under different operating conditions, and finally an implementation proposal of the developed project is presented. |
|---|