Avaliação da Estabilidade de Tensão Utilizando o Índice |D’| e Redes Neurais Artificiais sob Contingências
Approaches using Artificial Neural Networks (ANNs) have aimed to enhance the accuracy and reliability of voltage stability index calculations to ensure the secure operation of Power Systems (PS), especially under conditions of imminent voltage collapse. Furthermore, complementary research has incorp...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | Brasil |
| Institución: | Universidade Federal de Mato Grosso (UFMT) |
| Repositorio: | E&S Engineering and Science |
| Idioma: | portugués |
| OAI Identifier: | oai:periodicoscientificos.ufmt.br:article/18663 |
| Acceso en línea: | https://periodicoscientificos.ufmt.br/ojs/index.php/eng/article/view/18663 |
| Access Level: | acceso abierto |
| Palabra clave: | Estabilidade de Tensão Índice de Estabilidade de Tensão Redes Neurais Artificiais Voltage Stability Artificial Neural Networks Voltage Stability Index Estabilidad de Tensión. Redes Neuronales Artificiales Índice de Estabilidad de Tensión |
| Sumario: | Approaches using Artificial Neural Networks (ANNs) have aimed to enhance the accuracy and reliability of voltage stability index calculations to ensure the secure operation of Power Systems (PS), especially under conditions of imminent voltage collapse. Furthermore, complementary research has incorporated the dynamic modeling of transformers and renewable energy sources, while also leveraging real-time phasor measurements to enhance these methodologies. Despite significant advancements, there is still a need to improve the accuracy and computational efficiency of existing indices, particularly in multiple contingency scenarios. This paper proposes the use of the |D’| index, derived from the Power Flow Jacobian matrix, to enhance the precision of voltage stability assessment. The proposed method is evaluated through simulations considering three types of contingencies: stepwise increase in active and reactive power at loads, continuation power flow analysis, and transmission line outages. Performance tests of the |D’| index, using ANNs, demonstrate high accuracy and strong generalization, with low mean absolute errors and standard deviation values, enabling the efficient identification of the most critical buses in the system with low computational cost. The proposed method proved to be effective in reducing errors and variance during testing and validation, particularly under operating conditions close to voltage collapse, highlighting its robustness and efficiency in real-time stability analysis. |
|---|