แจ้งเอกสารไม่ครบถ้วน, ไม่ตรงกับชื่อเรื่อง หรือมีข้อผิดพลาดเกี่ยวกับเอกสาร ติดต่อที่นี่ ==>
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
Rakkrit Duangsoithong. Comparison of feature extraction methods for classifying energy theft and defective meters in automatic meter reading. (). King Mongkut's University of Technology North Bangkok. Central Library. : , 2023.
Comparison of feature extraction methods for classifying energy theft and defective meters in automatic meter reading
Name: Supakan Janthong
Organization :
Prince of Songkla University. Department of Electrical Engineering
Email :
Supakan.jan@pea.co.th
Name: Kusumal Chalermyanont
Organization :
Prince of Songkla University. Department of Electrical Engineering
Email :
Kusumal.c@.psu.ac.th
Name: Rakkrit Duangsoithong
Organization :
Prince of Songkla University. Department of Electrical Engineering
Email :
Rakkrit.d@.psu.ac.th
keyword:
Automatic Meter Reading (AMR)
ThaSH:
Electric power consumption
-- Control
Abstract:
Loss is an important factor for any organization
that operates in the distribution of electric power. Non-
Technical Loss (NTL) is one of the challenging losses. Most
distributors attempt to find a way to reduce the NTL loss. In
some cases, the anomaly patterns are complex, similar, and
difficult to classify. This paper proposes a feature extraction
method using signal modeling methods obtained from voltage
and current in a 3-phase system for abnormalities in the
Automatic Meter Reading (AMR) of the Provincial Electricity
Authority (PEA). Focusing on the main causes of NTL include
energy theft and defective meters using supervised learning for
classification. All-pole modeling has been applied to extract the
feature as a coefficient. The voltage and current load profiles are
divided into 1-week intervals (672 values) and fed into the signal
modeling to be extracted as 4 coefficients. The methods can
reduce the number of features before input into the model,
which makes the model process faster and improves
performance. The experimental results show that extraction
using Prony's method and k-Nearest Neighbors (kNN) model
significantly outperforms other methods.
King Mongkut's University of Technology North Bangkok. Central Library
Address:
BANGKOK
Email:
library@kmutnb.ac.th
Created:
2023
Modified:
2024-12-02
Issued:
2024-12-02
บทความ/Article
application/pdf
BibliograpyCitation :
In Electrical Engineering Academic Association (Thailand), Mahasarakham University. Faculty of Engineering and ASEFA. The 2023 International Electrical Engineering Congress (iEECON 2023) (pp.49-53). Mahasarakham : Mahasarakham University
eng
©copyrights King Mongkut's University of Technology North Bangkok
RightsAccess:
ใช้เวลา
0.029272 วินาที
Supakan Janthong
| Title | Contributor | Type |
|---|---|---|
| Comparison of feature extraction methods for classifying energy theft and defective meters in automatic meter reading
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Supakan Janthong;Kusumal Chalermyanont;Rakkrit Duangsoithong | บทความ/Article |
Kusumal Chalermyanont
| Title | Contributor | Type |
|---|---|---|
| Comparison of feature extraction methods for classifying energy theft and defective meters in automatic meter reading
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Supakan Janthong;Kusumal Chalermyanont;Rakkrit Duangsoithong | บทความ/Article |
Rakkrit Duangsoithong
| Title | Contributor | Type |
|---|---|---|
| Comparison of feature extraction methods for classifying energy theft and defective meters in automatic meter reading
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Supakan Janthong;Kusumal Chalermyanont;Rakkrit Duangsoithong | บทความ/Article |