Abstract
A complicated voltage fluctuation in distribution systems and a decline in power quality occur when a large number of photovoltaic (PV) systems are installed. In this paper, the installation of a low-voltage regulator is assumed, and a method for instantly and accurately determining the line drop compensator (LDC) parameters is proposed to perform efficient voltage management, which consists of prediction, operation, and control. In the proposed method, the computational cost to derive the LDC parameters can be reduced by learning the optimality of the parameters in a series of load demands and the PV output using multiple classifiers. We performed numerical simulations to verify the validity of the proposed method. From the results, the classification accuracy is found to improve by considering the majority vote of multiple classifiers. Additionally, the improvement in the voltage control performance is verified.
Original language | English |
---|---|
Title of host publication | IEEE PES Innovative Smart Grid Technologies Conference Europe |
Publisher | IEEE Computer Society |
Volume | 2015-January |
Edition | January |
DOIs | |
Publication status | Published - 2015 Jan 30 |
Event | 2014 IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014 - Istanbul, Turkey Duration: 2014 Oct 12 → 2014 Oct 15 |
Other
Other | 2014 IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2014 |
---|---|
Country | Turkey |
City | Istanbul |
Period | 14/10/12 → 14/10/15 |
Fingerprint
Keywords
- Distribution system
- K-nearest-neighbor algorithm
- LVR
- Random forests
- Support vector machine
ASJC Scopus subject areas
- Computer Networks and Communications
- Information Systems
Cite this
Method for instantly determining line drop compensator parameters of low-voltage regulator using multiple classifiers. / Kikusato, Hiroshi; Takahashi, Naoyuki; Yoshinaga, Jun; Fujimoto, Yu; Hayashi, Yasuhiro; Kusagawa, Shinichi; Motegi, Noriyuki.
IEEE PES Innovative Smart Grid Technologies Conference Europe. Vol. 2015-January January. ed. IEEE Computer Society, 2015. 7028851.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Method for instantly determining line drop compensator parameters of low-voltage regulator using multiple classifiers
AU - Kikusato, Hiroshi
AU - Takahashi, Naoyuki
AU - Yoshinaga, Jun
AU - Fujimoto, Yu
AU - Hayashi, Yasuhiro
AU - Kusagawa, Shinichi
AU - Motegi, Noriyuki
PY - 2015/1/30
Y1 - 2015/1/30
N2 - A complicated voltage fluctuation in distribution systems and a decline in power quality occur when a large number of photovoltaic (PV) systems are installed. In this paper, the installation of a low-voltage regulator is assumed, and a method for instantly and accurately determining the line drop compensator (LDC) parameters is proposed to perform efficient voltage management, which consists of prediction, operation, and control. In the proposed method, the computational cost to derive the LDC parameters can be reduced by learning the optimality of the parameters in a series of load demands and the PV output using multiple classifiers. We performed numerical simulations to verify the validity of the proposed method. From the results, the classification accuracy is found to improve by considering the majority vote of multiple classifiers. Additionally, the improvement in the voltage control performance is verified.
AB - A complicated voltage fluctuation in distribution systems and a decline in power quality occur when a large number of photovoltaic (PV) systems are installed. In this paper, the installation of a low-voltage regulator is assumed, and a method for instantly and accurately determining the line drop compensator (LDC) parameters is proposed to perform efficient voltage management, which consists of prediction, operation, and control. In the proposed method, the computational cost to derive the LDC parameters can be reduced by learning the optimality of the parameters in a series of load demands and the PV output using multiple classifiers. We performed numerical simulations to verify the validity of the proposed method. From the results, the classification accuracy is found to improve by considering the majority vote of multiple classifiers. Additionally, the improvement in the voltage control performance is verified.
KW - Distribution system
KW - K-nearest-neighbor algorithm
KW - LVR
KW - Random forests
KW - Support vector machine
UR - http://www.scopus.com/inward/record.url?scp=84936973076&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84936973076&partnerID=8YFLogxK
U2 - 10.1109/ISGTEurope.2014.7028851
DO - 10.1109/ISGTEurope.2014.7028851
M3 - Conference contribution
AN - SCOPUS:84936973076
VL - 2015-January
BT - IEEE PES Innovative Smart Grid Technologies Conference Europe
PB - IEEE Computer Society
ER -