Evaluating Generating Unit Unavailability Using Bayesian Power Priors

Generating unit unavailability assessment is an important task in the power system generation expansion planning aimed at managing an acceptable degree of security of supply. In the case of the newly installed or planned units, the sample size of the data for unavailability assessment is limited and...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on power systems 2017-05, Vol.32 (3), p.2315-2323
Main Authors: Lorencin, Ivan, Pantos, MiloS
Format: Article
Language:English
Subjects:
Accuracy
Availability
Bayes methods
Bayesian analysis
Bayesian inference
Confidence intervals
Empirical analysis
Generating unit reliability
informative prior
Maximum likelihood estimation
power generation
Power system reliability
Reliability
Standard error
Statistical analysis
unavailability
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Generating unit unavailability assessment is an important task in the power system generation expansion planning aimed at managing an acceptable degree of security of supply. In the case of the newly installed or planned units, the sample size of the data for unavailability assessment is limited and inadequate to provide the desired accuracy in the unavailability estimation. A new concept based on the Bayesian power prior approach has been developed to utilize the data from similar generating units. The original contribution of the present work is a model that incorporates the data of unavailability from other generating units into the statistical analysis of unavailability of the analysed generating unit to improve the accuracy of the estimation. The empirical results show that for unavailability estimation, the power prior Bayesian approach exhibits better than the classical statistical approach in both the standard error of estimate and confidence interval as the measures of accuracy.
ISSN:0885-8950
1558-0679
DOI:10.1109/TPWRS.2016.2603469