G. Arroyo-Figueroa (Mexico)
Diagnosis faults, uncertainty, Bayesian networks, temporal reasoning, intelligent systems, power systems.
The purpose of this paper is to show the relevance of Bayesian networks to model the uncertainty in some power systems problems. Uncertainty has been modelled based on randomness or stochastic models for random load variations, noise in measures for state estimation, fluctuations in model parameters, and so on. Some potential applications areas are contingency analysis, fault diagnosis, monitoring, distribution planning, generation dispatch, load forecasting, and load management. These uncertainties occur due to failures of protective delays and breakers, errors of local acquisition and transmission, and inaccurate occurrence time, etc. To deal with this kind of uncertainties, we proposed a novel approach based on a Temporal Bayesian network of events (TBNE). In a TBNE each node represents an event or state change and an arc corresponds to a causal-temporal relationship between the events. With this kind of knowledge representation for dealing uncertainty and time, we can diagnosis faults and estimated defects, and predict outcome events. Examples in the area of diagnostics are used to illustrate the types of uncertainties in power systems problems that are well represented by Bayesian networks and TBNE methods.
Important Links:
Go Back
