Machine learning-based ice detection approach for power transmission lines by utilizing FBG micro-meteorological...

Area of Science:

• Electrical Engineering
• Atmospheric Science
• Data Science

Background:

• Severe icing on transmission lines leads to power failures and infrastructure damage.
• Current icing monitoring techniques suffer from irregular data collection and low accuracy.

Purpose of the Study:

• To develop a comprehensive model for predicting hard rime and glaze ice on transmission lines.
• To improve the accuracy and reliability of icing monitoring systems.

Main Methods:

• A predictive model integrating temperature, humidity, and historical icing data was developed.
• The model's performance was validated through experimental verification across nine icing cycles.
• Data from different years and geographic locations were used for robust testing.

Main Results:

• The proposed model demonstrated high accuracy in predicting multiple types of ice accumulation.
• Achieved correlation coefficients greater than 0.99.
• Maintained a mean squared error below 4%.

Conclusions:

• The developed model offers an effective solution for predicting hard rime and glaze ice.
• This advancement enhances the reliability of transmission line operations and power grid stability.
• The technique provides a foundation for more accurate and consistent environmental hazard monitoring.