Intelligent compensation method for measurement errors in optical fiber current sensor caused by temperature...

Area of Science:

• Electrical Engineering
• Sensor Technology
• Artificial Intelligence

Background:

• Fiber optic current sensors (FOCS) are crucial for power systems but susceptible to temperature variations, impacting measurement accuracy.
• Accurate current sensing is vital for high-voltage transmission and renewable energy integration, where environmental conditions fluctuate.
• Existing compensation methods often require complex hardware modifications, limiting their applicability.

Purpose of the Study:

• To develop an intelligent error compensation method for FOCS to mitigate temperature-induced inaccuracies.
• To leverage easily measurable parameters for predicting and compensating temperature-dependent errors in FOCS.
• To validate the proposed method's effectiveness and robustness in harsh environmental conditions.

Main Methods:

• An improved Quantum-behaved Particle Swarm Optimization-Neural Network (Levy-Weighted-QPSO-NN) algorithm was developed.
• The algorithm uses sensing ring temperature, optical power, half-wave voltage, and SLD parameters as inputs.
• Experimental validation involved temperature cycling (-45 °C to 70 °C) on three sensing rings.

Main Results:

• The Levy-Weighted-QPSO-NN model achieved 91.11% average prediction accuracy for current ratio difference (R²=0.9223).
• The model outperformed standard QPSO-NN (85.69%) and Weighted-QPSO-NN (88.31%) algorithms.
• Real-time compensation reduced measurement errors from 0.82% to 0.13%, meeting Class 0.2S accuracy standards.

Conclusions:

• The Levy-Weighted-QPSO-NN algorithm provides a robust, algorithm-driven solution for temperature compensation in FOCS.
• This method enhances FOCS accuracy and stability without requiring hardware modifications.
• The approach offers a generic solution for improving FOCS performance in critical power applications.