Sensor Fusion for Power Line Sensitive Monitoring and Load State Estimation

Area of Science:

• Electrical Engineering
• Control Systems
• Signal Processing

Background:

• Transformer systems are critical in power grids, and their reliable operation depends on effective fault detection.
• Traditional fault detection methods may require extensive sensor installations or lack precision in identifying specific fault types.

Purpose of the Study:

• To develop and validate advanced fault detection techniques for transformer systems.
• To utilize the extended Kalman filter (EKF) for soft sensing and fault identification using limited primary measurements.

Main Methods:

• Two distinct EKFs were designed: one for estimating secondary winding parameters and detecting short circuits, and another for harmonic analysis of primary voltage.
• The transformer model based on mutual inductance was employed for observer design.
• Simulations were conducted to evaluate the performance of the proposed EKF observers.

Main Results:

• The first EKF successfully detected short circuits in the secondary winding by estimating secondary voltage, current, and load resistance.
• The second EKF accurately estimated the amplitude and frequency of primary voltage harmonics.
• The proposed methods demonstrated efficiency in fault detection and soft sensing capabilities.

Conclusions:

• EKF observers provide a robust mathematical framework for fault detection in transformer systems.
• These methods are valuable for sensor fusion, integrating diverse data for accurate system monitoring.
• The proposed approach enhances the safety and reliability of power systems through advanced fault diagnosis.