Multivariate fault isolation methods based on penalized regression are proposed.
•
The problem is transformed to a variable selection task for discriminant analysis.
•
The variable selection problem for DA is further linked to penalized regression.
•
The smearing effect is avoided.
•
Historical fault data or fault direction information is not required.
NGLC 2004-2010.National Geological Library of China All Rights Reserved.
Add:29 Xueyuan Rd,Haidian District,Beijing,PRC. Mail Add: 8324 mailbox 100083
For exchange or info please contact us via email.