학술논문

A novel method for constructing the Bayesian Network for detailed reliability assessment of Power Systems
Document Type
Conference
Source
2009 International Conference on Electric Power and Energy Conversion Systems, (EPECS) Electric Power and Energy Conversion Systems, 2009. EPECS '09. International Conference on. :1-6 Nov, 2009
Subject
Geoscience
Power, Energy and Industry Applications
Power system reliability
Bayesian methods
Power system analysis computing
Power system simulation
Power system modeling
Power system planning
Computer network reliability
Training data
System testing
Inference algorithms
Bayesian Network
Power System
Reliability
Language
Abstract
In this paper, a novel approach is presented for constructing the Bayesian Network (BN) that can be used in detailed reliability assessments of a power system based on the loss of load probability (LOLP). This method, which is applicable to large and complex power systems, is based on the learning capability of BN from data. In order to obtain training data, state sampling method of the Monte Carlo Simulation is used. The proposed method is applied to two test systems and various analyses are performed showing its efficiency and validity.