Enhancing the distribution power system resilience against hurricane events using a bayesian network line outage prediction model
Abstract Enhancing the grid resilience against hurricane events proactively requires a pre‐disaster system optimization built on accurate system network line outage prediction. In the past, the statistical system level failure predictive model such as generalized linear model (GLM), generalized addi...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Wiley
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/d3ba041a47ef4fdf9631e6ab06e04240 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Abstract Enhancing the grid resilience against hurricane events proactively requires a pre‐disaster system optimization built on accurate system network line outage prediction. In the past, the statistical system level failure predictive model such as generalized linear model (GLM), generalized additive model (GAM), system tree‐based mining model (classification regression tree (CART), Bayesian additive regression model (BART), and the topology‐based system components’ fragility curve (FC)‐Monte Carlo simulation (MCS) model have been used to estimate the hurricane‐induced damage on the grid system. Although these models are suitable for a long term infrastructural planning, increased prediction approximation error with high computational time limit their applications as a components level predictive model that can be used for a short‐term proactive operational planning. To solve these problems, a dynamic bayesian network (BN) model is proposed. The investigation is performed on a standard IEEE 15‐bus system using hurricane events data. The proposed BN's system line outage prediction accuracy and efficiency are validated using the statistical system grid components’ FC‐MCS‐scenario reduction (SCENRED) predictive model. Therefore, creating a platform to develop a cost‐effective pre‐disaster system optimization for system network resilience enhancement against the predicted approaching hurricane events. |
|---|