ISSUES BY YEAR
Volume 12 - 2016
Volume 11 - 2015
Volume 10 - 2014
Volume 9 - 2013
Volume 8 - 2012
Volume 7 - 2011
Volume 6 - 2010
Volume 5 - 2009
Volume 4 - 2008
Volume 3 - 2007
Volume 2 - 2006
|Written by P. BARALDI, M. COMPARE, E. ZIO, M. DE NIGRIS, G. RIZZI|
Identification of Contradictory Patterns in Experimental Datasets for the Development of Models for Electrical Cables Diagnostics
Volume 7, Number 1, January 2011 - Paper 4 - pp. 43-60
P. BARALDI1, M. COMPARE1, E. ZIO1,2, M. DE NIGRIS3, G. RIZZI31 Energy Department, Politecnico di Milano, Milan, Italy
2 Ecole Centrale Paris-Supelec, Paris, France
3 Enea Ricerca sul Sistema Elettrico – ERSE, Milan, Italy
(Received on March 30, 2010, revised on August 1, 2010)
The state of health of an electrical cable may be difficult to know, without destructive or very expensive tests. To overcome this, partial discharge (PD) measurements have been proposed as a relatively economic and simple-to-apply experimental technique for retrieving information on the state of health of an electrical cable. The retrieval is based on a relationship between PD measurements and the health of the cable. Given the difficulties in capturing such relationship by analytical models, empirical modeling techniques based on experimental data have been propounded. In this view, a set of PD measurements have been collected by Enea Ricerca sul Sistema Elettrico-ERSE during past campaigns, for building a diagnostic system of electrical cable health state. These experimental data may contain contradictory information which remarkably reduces the performance of the state classifier, if not a priori identified and possibly corrected. In the present paper, a novel technique based on the Adaboost algorithm is proposed for identifying contradictory PD patterns within an a priori analysis aimed at improving the diagnostic performance. Adaboost is a bootstrap-inspired, ensemble-based algorithm which has been effectively used for addressing classification problems.
Click here to download the paper.
Please note : You will need Adobe Acrobat viewer to view the full articles.
Important Access Information
Individuals, Institutions and Corporations with access via userid and password:
If you have a valid userid and password, you will need to login at the top of the screen. All volumes that are authorized by your subscription will be available for download. If you are not authorized, you will see an 'add to cart' option and you have the choice of purchasing the articles. You may also apply for subscription from our Subscription page.
If you cannot access any paper and you feel your subscription entitles you to access, please notify us by using the contact form on the Contact Us page. Please provide as much detail as possible.
Institutions and Corporations with access via IP addresses:
If you have a subscription via IP addresses, all volumes that are authorized by your subscription will be available for download. If you are not authorized, you will see an 'add to cart' option and you have the choice of purchasing the articles. You may also apply for subscription from our Subscription page.
If you cannot access any paper and you feel your subscription entitles you to access, please verify that the EXTERNAL IP address of your computer is authorized. Before contacting us, contact your system administrator and confirm if the IP address of your network has been authorized for access. To find out the EXTERNAL IP address of your network, simply open any Internet browser and point it to : http://www.WhatIsMyIP.com. Your EXTERNAL IP address will be shown in the browser window.
Once you have this information, you may email us using the contact form on the Contact Us page. Please provide as much detail as possible. You may also email a screenshot of the browser to :subscriptions@IJPE-Online.com
Thank you for your understanding. We will try our best to reply within 24-48 hours.