Statistical methods for condition assessment of low-failure assets

Maikel Klerx; Johan Morren; Han Slootweg

doi:10.1109/PTC.2019.8810966

Statistical methods for condition assessment of low-failure assets

Maikel Klerx, Johan Morren, Han Slootweg

Electrical Energy Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

For (electrical) infrastructure assets a high reliability is required. When assets cannot be monitored and no condition information is available, statistical methods can be used to assess their condition. This paper discusses statistical methods to assess the condition of LV cables, which is a typical example of electric components with a very low failure rate. Two survival methods and three classification methods are compared. The Cox Proportional Hazard model proved to be the most suitable. The influence of variables on the failure probability provides insight in failure mechanisms and asset groups which have a high failure probability. In this paper it is proved that a great amount of information can be gained using survival analysis for ageing infrastructure assets.

Original language	English
Title of host publication	2019 IEEE Milan PowerTech, PowerTech 2019
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Number of pages	6
ISBN (Electronic)	978-1-5386-4722-6
DOIs	https://doi.org/10.1109/PTC.2019.8810966
Publication status	Published - 1 Jun 2019
Event	2019 IEEE Milan PowerTech, PowerTech 2019 - Milan, Italy Duration: 23 Jun 2019 → 27 Jun 2019

Conference

Conference	2019 IEEE Milan PowerTech, PowerTech 2019
Country	Italy
City	Milan
Period	23/06/19 → 27/06/19

Keywords

Asset management
Big data applications
Power distribution
Power grids
Regression analysis

Access to Document

10.1109/PTC.2019.8810966

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Statistical methods for condition assessment of low-failure assets'. Together they form a unique fingerprint.

Cite this

@inproceedings{4152616f128440b2850ebddc901d633a,

title = "Statistical methods for condition assessment of low-failure assets",

abstract = "For (electrical) infrastructure assets a high reliability is required. When assets cannot be monitored and no condition information is available, statistical methods can be used to assess their condition. This paper discusses statistical methods to assess the condition of LV cables, which is a typical example of electric components with a very low failure rate. Two survival methods and three classification methods are compared. The Cox Proportional Hazard model proved to be the most suitable. The influence of variables on the failure probability provides insight in failure mechanisms and asset groups which have a high failure probability. In this paper it is proved that a great amount of information can be gained using survival analysis for ageing infrastructure assets.",

keywords = "Asset management, Big data applications, Power distribution, Power grids, Regression analysis",

author = "Maikel Klerx and Johan Morren and Han Slootweg",

year = "2019",

month = jun,

day = "1",

doi = "10.1109/PTC.2019.8810966",

language = "English",

booktitle = "2019 IEEE Milan PowerTech, PowerTech 2019",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2019 IEEE Milan PowerTech, PowerTech 2019 ; Conference date: 23-06-2019 Through 27-06-2019",

}

TY - GEN

T1 - Statistical methods for condition assessment of low-failure assets

AU - Klerx, Maikel

AU - Morren, Johan

AU - Slootweg, Han

PY - 2019/6/1

Y1 - 2019/6/1

N2 - For (electrical) infrastructure assets a high reliability is required. When assets cannot be monitored and no condition information is available, statistical methods can be used to assess their condition. This paper discusses statistical methods to assess the condition of LV cables, which is a typical example of electric components with a very low failure rate. Two survival methods and three classification methods are compared. The Cox Proportional Hazard model proved to be the most suitable. The influence of variables on the failure probability provides insight in failure mechanisms and asset groups which have a high failure probability. In this paper it is proved that a great amount of information can be gained using survival analysis for ageing infrastructure assets.

AB - For (electrical) infrastructure assets a high reliability is required. When assets cannot be monitored and no condition information is available, statistical methods can be used to assess their condition. This paper discusses statistical methods to assess the condition of LV cables, which is a typical example of electric components with a very low failure rate. Two survival methods and three classification methods are compared. The Cox Proportional Hazard model proved to be the most suitable. The influence of variables on the failure probability provides insight in failure mechanisms and asset groups which have a high failure probability. In this paper it is proved that a great amount of information can be gained using survival analysis for ageing infrastructure assets.

KW - Asset management

KW - Big data applications

KW - Power distribution

KW - Power grids

KW - Regression analysis

UR - http://www.scopus.com/inward/record.url?scp=85072344569&partnerID=8YFLogxK

U2 - 10.1109/PTC.2019.8810966

DO - 10.1109/PTC.2019.8810966

M3 - Conference contribution

BT - 2019 IEEE Milan PowerTech, PowerTech 2019

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 2019 IEEE Milan PowerTech, PowerTech 2019

Y2 - 23 June 2019 through 27 June 2019

ER -