Surge arrester leakage current modeling based on pollution layer electrical conductivity estimation.

Area of Science:

• Electrical Engineering
• Materials Science
• Environmental Science

Background:

• Surface leakage current (LC) in metal oxide surge arresters (MOSA) is significantly affected by environmental factors.
• Accurate modeling of MOSA LC is crucial for reliable performance and as an alternative to extensive laboratory testing.

Purpose of the Study:

• To develop a new method for modeling MOSA surface LC by estimating the electrical conductivity (EC) of the contaminated layer.
• To accurately predict MOSA LC under diverse environmental conditions.

Main Methods:

• Utilized artificial intelligence (AI) to estimate the EC of polluted layers on silicon rubber surge arresters.
• Investigated the impact of uniform/non-uniform pollution, humidity, pollution intensity, and voltage.
• Employed Finite Element Method (FEM) software for surge arrester modeling using estimated EC.

Main Results:

• The AI-based method accurately estimated the EC of the contaminated layer, validated by Mean Squared Error (MSE) and Coefficient of Determination.
• The FEM model, incorporating estimated EC, effectively evaluated internal and external currents influencing total MOSA LC.
• The proposed model demonstrated capability in estimating EC and modeling LC, generalizing to conditions without prior lab data.

Conclusions:

• The developed AI-driven approach provides a highly accurate method for modeling MOSA surface LC under various environmental conditions.
• This technique offers a reliable alternative to laboratory testing for predicting MOSA performance.
• The study highlights the potential for generalizing the model to new scenarios, enhancing surge arrester reliability.