The application of machine learning for the voltage and reactive power control in power distribution network

Name: Phavixa Vongvilasack

Organization : Chiang Mai University. Faculty of Engineering

Email : Phavixa_cmu.ac.th

Name: Suttichai Premrudeepreechacharn

Organization : Chiang Mai University. Faculty of Engineering

Email : Suttic@gmail.com

Name: Kanchit Ngamsanroaj

Organization : Electricity Generating Authority of Thailand (EGAT)

Email : Kanchit@egat.co.th

keyword: Reactive power control

LCSH: Electric power distribution

; Particle swarm optimization

LCSH: Voltage-controlled oscillators

LCSH: Machine learning

LCSH: Distributed generation of electric power

LCSH: Support vector machines

LCSH: Decision trees

LCSH: Voltage regulators

Abstract: Recently, the integration of renewable energy sources into distribution systems has significantly affected the operation of conventional voltage regulation systems, requiring additional control actions. This study presents the application of machine learning (ML) for voltage and reactive power control in power distribution systems with the presence of distributed generators (DGs). The study aims to validate the effectiveness of ML algorithms in creating predictive models for pre-defining the coordinated operation of reactive power compensators for voltage regulation. An hourly dataset collected over three months from the Electricité du Laos (EDL) Savannakhet branch was applied to a simple radial feeder in the distribution network. Particle swarm optimization (PSO) was used to determine the optimal operating schedule of switched capacitor banks, and the control scheme was assumed to be based on centralized management for communication with reactive power support devices. A new dataset was created from the reactive power control process to provide knowledge for training ML algorithms. The algorithms utilized included Decision Trees, Support Vector Machine (SVM), and k-Nearest Neighbors, with feature selection also considered. The simulation results demonstrated that the ML classification algorithms provided satisfactory accuracy and improved voltage regulation performance. The developed model can assist system operators in monitoring and decision-making for system control

King Mongkut's University of Technology North Bangkok. Central Library

Address: BANGKOK

Email: library@kmutnb.ac.th

Created: 2022

Modified: 2026-05-14

Issued: 2026-05-14

บทความ/Article

application/pdf

BibliograpyCitation : In Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI) Association, Thailand. 2022 19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON 2022) (1570794149). Bangkok : ECTI Association

eng

©copyrights King Mongkut's University of Technology North Bangkok

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	ECTI-CON 2022 1570794149.pdf	2.01 MB

ใช้เวลา

0.032932 วินาที