Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems

Nontlahla May; Lutendo Muremi; Pitshou Bokoro; Siyasanga Innocent May; Wisani Hlaluku Mkasi

doi:10.1109/SAUPEC65723.2025.10944477

Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems

Nontlahla May
, Lutendo Muremi
, Pitshou Bokoro
, Siyasanga Innocent May
, Wisani Hlaluku Mkasi

Electrical and Electronic Engineering Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This research investigates the effectiveness of machine learning and deep learning models in forecasting voltage swell peak amplitudes within grid-connected photovoltaic (PV) systems, aiming to enhance power quality management. A 24-month dataset (January 2022 - December 2023) encompassing power and weather data from a 3.3 kWp PV system at the Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa, was utilized. Hourly averaged data between 5 am and 6 pm, capturing PV system and weather measurements, was analysed. Five models - Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short-Term Memory (LSTM) - were trained and evaluated using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). The Random Forest model demonstrated superior predictive accuracy, closely aligning with actual peak voltages and achieving the lowest MSE (0.01V²) and RMSE (0.02V). This study highlights the potential of machine learning, particularly Random Forest, in accurately predicting voltage swell peak amplitudes, contributing to improved power quality management in grid-connected PV systems.

Original language	English
Title of host publication	Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331535162
DOIs	https://doi.org/10.1109/SAUPEC65723.2025.10944477
Publication status	Published - 2025
Event	33rd Southern African Universities Power Engineering Conference, SAUPEC 2025 - Pretoria, South Africa Duration: 29 Jan 2025 → 30 Jan 2025

Publication series

Name	Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025

Conference

Conference	33rd Southern African Universities Power Engineering Conference, SAUPEC 2025
Country/Territory	South Africa
City	Pretoria
Period	29/01/25 → 30/01/25

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Grid-tied photovoltaic system
Machine learning algorithms
Peak amplitude prediction
Power quality
Voltage swell

ASJC Scopus subject areas

Artificial Intelligence
Energy Engineering and Power Technology
Renewable Energy, Sustainability and the Environment
Electrical and Electronic Engineering
Modeling and Simulation

Access to Document

10.1109/SAUPEC65723.2025.10944477

Cite this

May, N., Muremi, L., Bokoro, P., May, S. I., & Hlaluku Mkasi, W. (2025). Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems. In Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025 (Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SAUPEC65723.2025.10944477

May, Nontlahla ; Muremi, Lutendo ; Bokoro, Pitshou et al. / Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems. Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025).

@inproceedings{b93cc71d70b0489a8db83a07b85d8a6a,

title = "Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems",

abstract = "This research investigates the effectiveness of machine learning and deep learning models in forecasting voltage swell peak amplitudes within grid-connected photovoltaic (PV) systems, aiming to enhance power quality management. A 24-month dataset (January 2022 - December 2023) encompassing power and weather data from a 3.3 kWp PV system at the Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa, was utilized. Hourly averaged data between 5 am and 6 pm, capturing PV system and weather measurements, was analysed. Five models - Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short-Term Memory (LSTM) - were trained and evaluated using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). The Random Forest model demonstrated superior predictive accuracy, closely aligning with actual peak voltages and achieving the lowest MSE (0.01V2) and RMSE (0.02V). This study highlights the potential of machine learning, particularly Random Forest, in accurately predicting voltage swell peak amplitudes, contributing to improved power quality management in grid-connected PV systems.",

keywords = "Grid-tied photovoltaic system, Machine learning algorithms, Peak amplitude prediction, Power quality, Voltage swell",

author = "Nontlahla May and Lutendo Muremi and Pitshou Bokoro and May, \{Siyasanga Innocent\} and \{Hlaluku Mkasi\}, Wisani",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025 ; Conference date: 29-01-2025 Through 30-01-2025",

year = "2025",

doi = "10.1109/SAUPEC65723.2025.10944477",

language = "English",

series = "Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025",

address = "United States",

}

May, N, Muremi, L , Bokoro, P, May, SI & Hlaluku Mkasi, W 2025, Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems. in Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025. Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025, Institute of Electrical and Electronics Engineers Inc., 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025, Pretoria, South Africa, 29/01/25. https://doi.org/10.1109/SAUPEC65723.2025.10944477

Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems. / May, Nontlahla; Muremi, Lutendo ; Bokoro, Pitshou et al.
Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems

AU - May, Nontlahla

AU - Muremi, Lutendo

AU - Bokoro, Pitshou

AU - May, Siyasanga Innocent

AU - Hlaluku Mkasi, Wisani

PY - 2025

Y1 - 2025

N2 - This research investigates the effectiveness of machine learning and deep learning models in forecasting voltage swell peak amplitudes within grid-connected photovoltaic (PV) systems, aiming to enhance power quality management. A 24-month dataset (January 2022 - December 2023) encompassing power and weather data from a 3.3 kWp PV system at the Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa, was utilized. Hourly averaged data between 5 am and 6 pm, capturing PV system and weather measurements, was analysed. Five models - Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short-Term Memory (LSTM) - were trained and evaluated using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). The Random Forest model demonstrated superior predictive accuracy, closely aligning with actual peak voltages and achieving the lowest MSE (0.01V2) and RMSE (0.02V). This study highlights the potential of machine learning, particularly Random Forest, in accurately predicting voltage swell peak amplitudes, contributing to improved power quality management in grid-connected PV systems.

AB - This research investigates the effectiveness of machine learning and deep learning models in forecasting voltage swell peak amplitudes within grid-connected photovoltaic (PV) systems, aiming to enhance power quality management. A 24-month dataset (January 2022 - December 2023) encompassing power and weather data from a 3.3 kWp PV system at the Council for Scientific and Industrial Research (CSIR) in Pretoria, South Africa, was utilized. Hourly averaged data between 5 am and 6 pm, capturing PV system and weather measurements, was analysed. Five models - Artificial Neural Network (ANN), Support Vector Regression (SVR), Random Forest (RF), K-Nearest Neighbours (KNN), and Long Short-Term Memory (LSTM) - were trained and evaluated using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). The Random Forest model demonstrated superior predictive accuracy, closely aligning with actual peak voltages and achieving the lowest MSE (0.01V2) and RMSE (0.02V). This study highlights the potential of machine learning, particularly Random Forest, in accurately predicting voltage swell peak amplitudes, contributing to improved power quality management in grid-connected PV systems.

KW - Grid-tied photovoltaic system

KW - Machine learning algorithms

KW - Peak amplitude prediction

KW - Power quality

KW - Voltage swell

UR - https://www.scopus.com/pages/publications/105002679924

U2 - 10.1109/SAUPEC65723.2025.10944477

DO - 10.1109/SAUPEC65723.2025.10944477

M3 - Conference contribution

AN - SCOPUS:105002679924

T3 - Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025

BT - Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025

Y2 - 29 January 2025 through 30 January 2025

ER -

May N, Muremi L , Bokoro P, May SI, Hlaluku Mkasi W. Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems. In Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025. Institute of Electrical and Electronics Engineers Inc. 2025. (Proceedings of the 33rd Southern African Universities Power Engineering Conference, SAUPEC 2025). doi: 10.1109/SAUPEC65723.2025.10944477

Performance Evaluation of Machine Learning Models for Predicting Voltage Swell Peak Amplitude in Grid-tied Photovoltaic Systems

Abstract

Publication series

Conference

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this