<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In the quest for more secure power grids, this paper delves into the vital role of power monitoring systems and the burgeoning field of safety risk prediction. Traditional prediction methodologies falter due to slow computation and lackluster accuracy. Enter the XGBoost algorithm, hailed for its stellar performance in various prediction scenarios, yet still ripe for improvement within complex power system data. By marrying Operational Technology (OT) with Information Technology (IT), we elevate the predictive prowess of the XGBoost model. Our investigation, grounded in the analysis of 900 sample datasets, unveils a model with enhanced precision in security risk evaluation. This refined model not only surpasses traditional XGBoost in accuracy—with increased instances of near-perfect predictions—but also excels in vital statistical measures: reducing Mean Absolute Percentage Error (MAPE), lowering Root Mean Square Error (RMSE), and boosting both prediction stability and sensitivity. The introduction of the WOA-XGBoost algorithm marks a significant leap forward in fortifying power monitoring systems’ security and predictive alertness.</p>
</abstract>

In the quest for more secure power grids, this paper delves into the vital role of power monitoring systems and the burgeoning field of safety risk prediction. Traditional prediction methodologies falter due to slow computation and lackluster accuracy. Enter the XGBoost algorithm, hailed for its stellar performance in various prediction scenarios, yet still ripe for improvement within complex power system data. By marrying Operational Technology (OT) with Information Technology (IT), we elevate the predictive prowess of the XGBoost model. Our investigation, grounded in the analysis of 900 sample datasets, unveils a model with enhanced precision in security risk evaluation. This refined model not only surpasses traditional XGBoost in accuracy—with increased instances of near-perfect predictions—but also excels in vital statistical measures: reducing Mean Absolute Percentage Error (MAPE), lowering Root Mean Square Error (RMSE), and boosting both prediction stability and sensitivity. The introduction of the WOA-XGBoost algorithm marks a significant leap forward in fortifying power monitoring systems’ security and predictive alertness.

Research on Security Risk Prediction Technology of Electric Power Monitoring System under OT and IT Convergence

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In the quest for more secure power grids, this paper delves into the vital role of power monitoring systems and the burgeoning field of safety risk prediction....

Research on Security Risk Prediction Technology of Electric Power Monitoring System under OT and IT Convergence

Data publikacji: 01 kwi 2024

Zakres stron: -

Otrzymano: 02 lut 2024

Przyjęty: 07 lut 2024

DOI: https://doi.org/10.2478/amns-2024-0808

Słowa kluczowe
Power monitoring system, Security risk prediction, XGBoost algorithm, Machine learning

© 2024 Zhongfeng Wei et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Research on Security Risk Prediction Technology of Electric Power Monitoring System under OT and IT Convergence

Data publikacji: 01 kwi 2024

Zakres stron: -

Otrzymano: 02 lut 2024

Przyjęty: 07 lut 2024

DOI: https://doi.org/10.2478/amns-2024-0808

Słowa kluczowePower monitoring system, Security risk prediction, XGBoost algorithm, Machine learning

© 2024 Zhongfeng Wei et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Słowa kluczowe
Power monitoring system, Security risk prediction, XGBoost algorithm, Machine learning