Abstract
Power companies can benefit from the use of knowledge discovery methods and statistical machine learning for preventive maintenance. We introduce a general process for transforming historical electrical grid data into models that aim to predict the risk of failures for components and systems. These models can be used directly by power companies to assist with prioritization of maintenance and repair work. Specialized versions of this process are used to produce 1) feeder failure rankings, 2) cable, joint, terminator, and transformer rankings, 3) feeder Mean Time Between Failure (MTBF) estimates, and 4) manhole events vulnerability rankings. The process in its most general form can handle diverse, noisy, sources that are historical (static), semi-real-time, or real-time, incorporates state-of-the-art machine learning algorithms for prioritization (supervised ranking or MTBF), and includes an evaluation of results via cross-validation and blind test. Above and beyond the ranked lists and MTBF estimates are business management interfaces that allow the prediction capability to be integrated directly into corporate planning and decision support; such interfaces rely on several important properties of our general modeling approach: that machine learning features are meaningful to domain experts, that the processing of data is transparent, and that prediction results are accurate enough to support sound decision making. We discuss the challenges in working with historical electrical grid data that were not designed for predictive purposes. The rawness of these data contrasts with the accuracy of the statistical models that can be obtained from the process; these models are sufficiently accurate to assist in maintaining New York City's electrical grid.
| Original language | English (US) |
|---|---|
| Article number | 5770269 |
| Pages (from-to) | 328-345 |
| Number of pages | 18 |
| Journal | IEEE Transactions on Pattern Analysis and Machine Intelligence |
| Volume | 34 |
| Issue number | 2 |
| DOIs | |
| State | Published - Jan 1 2012 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Software
- Computer Vision and Pattern Recognition
- Computational Theory and Mathematics
- Artificial Intelligence
- Applied Mathematics
Cite this
}
Machine learning for the New York City power grid. / Rudin, Cynthia; Waltz, David; Anderson, Roger; Boulanger, Albert; Salleb-Aouissi, Ansaf; Chow, Maggie; Dutta, Haimonti; Gross, Philip; Huang, Bert; Ierome, Steve; Isaac, Delfina F.; Kressner, Arthur; Passonneau, Rebecca Jane; Radeva, Axinia; Wu, Leon.
In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 34, No. 2, 5770269, 01.01.2012, p. 328-345.Research output: Contribution to journal › Article
TY - JOUR
T1 - Machine learning for the New York City power grid
AU - Rudin, Cynthia
AU - Waltz, David
AU - Anderson, Roger
AU - Boulanger, Albert
AU - Salleb-Aouissi, Ansaf
AU - Chow, Maggie
AU - Dutta, Haimonti
AU - Gross, Philip
AU - Huang, Bert
AU - Ierome, Steve
AU - Isaac, Delfina F.
AU - Kressner, Arthur
AU - Passonneau, Rebecca Jane
AU - Radeva, Axinia
AU - Wu, Leon
PY - 2012/1/1
Y1 - 2012/1/1
N2 - Power companies can benefit from the use of knowledge discovery methods and statistical machine learning for preventive maintenance. We introduce a general process for transforming historical electrical grid data into models that aim to predict the risk of failures for components and systems. These models can be used directly by power companies to assist with prioritization of maintenance and repair work. Specialized versions of this process are used to produce 1) feeder failure rankings, 2) cable, joint, terminator, and transformer rankings, 3) feeder Mean Time Between Failure (MTBF) estimates, and 4) manhole events vulnerability rankings. The process in its most general form can handle diverse, noisy, sources that are historical (static), semi-real-time, or real-time, incorporates state-of-the-art machine learning algorithms for prioritization (supervised ranking or MTBF), and includes an evaluation of results via cross-validation and blind test. Above and beyond the ranked lists and MTBF estimates are business management interfaces that allow the prediction capability to be integrated directly into corporate planning and decision support; such interfaces rely on several important properties of our general modeling approach: that machine learning features are meaningful to domain experts, that the processing of data is transparent, and that prediction results are accurate enough to support sound decision making. We discuss the challenges in working with historical electrical grid data that were not designed for predictive purposes. The rawness of these data contrasts with the accuracy of the statistical models that can be obtained from the process; these models are sufficiently accurate to assist in maintaining New York City's electrical grid.
AB - Power companies can benefit from the use of knowledge discovery methods and statistical machine learning for preventive maintenance. We introduce a general process for transforming historical electrical grid data into models that aim to predict the risk of failures for components and systems. These models can be used directly by power companies to assist with prioritization of maintenance and repair work. Specialized versions of this process are used to produce 1) feeder failure rankings, 2) cable, joint, terminator, and transformer rankings, 3) feeder Mean Time Between Failure (MTBF) estimates, and 4) manhole events vulnerability rankings. The process in its most general form can handle diverse, noisy, sources that are historical (static), semi-real-time, or real-time, incorporates state-of-the-art machine learning algorithms for prioritization (supervised ranking or MTBF), and includes an evaluation of results via cross-validation and blind test. Above and beyond the ranked lists and MTBF estimates are business management interfaces that allow the prediction capability to be integrated directly into corporate planning and decision support; such interfaces rely on several important properties of our general modeling approach: that machine learning features are meaningful to domain experts, that the processing of data is transparent, and that prediction results are accurate enough to support sound decision making. We discuss the challenges in working with historical electrical grid data that were not designed for predictive purposes. The rawness of these data contrasts with the accuracy of the statistical models that can be obtained from the process; these models are sufficiently accurate to assist in maintaining New York City's electrical grid.
UR - http://www.scopus.com/inward/record.url?scp=84055217877&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84055217877&partnerID=8YFLogxK
U2 - 10.1109/TPAMI.2011.108
DO - 10.1109/TPAMI.2011.108
M3 - Article
C2 - 21576741
AN - SCOPUS:84055217877
VL - 34
SP - 328
EP - 345
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
SN - 0162-8828
IS - 2
M1 - 5770269
ER -