2006 IEEE PES Power Systems Conference and Exposition 2006
DOI: 10.1109/psce.2006.296481
|View full text |Cite
|
Sign up to set email alerts
|

Data Mining and Analysis of Tree-Caused Faults in Power Distribution Systems

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
11
0

Year Published

2006
2006
2018
2018

Publication Types

Select...
4
4

Relationship

2
6

Authors

Journals

citations
Cited by 29 publications
(11 citation statements)
references
References 3 publications
0
11
0
Order By: Relevance
“…Each outage record in this database contains 33 information fields. Based on Duke Energy senior distribution engineers' suggestions as well as the statistical significance tests [7], six of the fields are considered as the most essential and influential factors: circuit ID, weather, season, time of day, phases affected, and protective devices activated. These six influential factors are chosen to be independent variables, and the attribute cause entered by the crew after finding out the actual outage cause during the restoration process is selected as the dependent variable (i.e., the class label).…”
Section: Duke Energy Outage Data and Data Preprocessingmentioning
confidence: 99%
See 1 more Smart Citation
“…Each outage record in this database contains 33 information fields. Based on Duke Energy senior distribution engineers' suggestions as well as the statistical significance tests [7], six of the fields are considered as the most essential and influential factors: circuit ID, weather, season, time of day, phases affected, and protective devices activated. These six influential factors are chosen to be independent variables, and the attribute cause entered by the crew after finding out the actual outage cause during the restoration process is selected as the dependent variable (i.e., the class label).…”
Section: Duke Energy Outage Data and Data Preprocessingmentioning
confidence: 99%
“…All six independent variables are categorical as shown in Table I; they are transformed into numerical variables using the likelihood measure [7] so that they can be used in most of the commonly used classification processes, including ANN and AIRS, which usually require numerical inputs.…”
Section: Duke Energy Outage Data and Data Preprocessingmentioning
confidence: 99%
“…Better understanding of causes and the associated consequences of faults will help to identify weak parts. This is useful for maintaining and operating the distribution network, as well as for developing the future system in terms of reliability considerations [5].…”
Section: A Problem Statementmentioning
confidence: 99%
“…This failure modeling is categorized as prediction problems, which can be solved by either parametric methods such as multivariable linear/exponential regressions or non-parametric methods such as artificial neural networks (ANNs) [4]. A few attempts have been made to evaluate failures as a function of either external factors, such as plant growth [4,5], weather [6], combination of features [7,8,9], or intrinsic factors, like age [10]. Reference [11] addresses a method that uses equipment inspection data to assign relative condition ranking, and [12] introduced a model, by categorizing the component failure rates into several partial failure rates.…”
Section: Introductionmentioning
confidence: 99%
“…Each outage record in this database contains 33 information fields. Based on Duke Energy senior distribution engineers' suggestions as well as the statistical significance tests [11], six of the fields are considered as the most essential and influential factors: circuit ID, weather, season, time of day, phases affected, and protective devices activated. All these six variables are categorical; they are transformed into numerical variables using the likelihood measure [11] so that they can be used in most of classification processes which usually require numerical inputs, including FAIRS, AIRS and E-algorithm.…”
Section: Duke Energy Outage Data and Data Preprocessingmentioning
confidence: 99%