What question did this study set out to answer?

To develop an optimal probabilistic neural network classifier to reliably distinguish between magnetizing inrush currents and internal faults in power transformer differential protection.

December 11, 2009IEEE Transactions on Power Delivery

Power Transformer Differential Protection Based On Optimal Probabilistic Neural Network

View Full Paper

Ask AI

Bookmark

Authors

MTManoj TripathyIndian Institute of Technology Roorkee RMR. P. MaheshwariIndian Institute of Technology Roorkee HVHarish VermaInstitute of Chartered Financial Analysts of India University, Jaipur

Discussion

Member takes

Implication

Simulation study demonstrates accurate discrimination between magnetizing inrush and internal faults in power transformers, indicating improved differential relay reliability.

Key Points

To develop an optimal probabilistic neural network classifier to reliably distinguish between magnetizing inrush currents and internal faults in power transformer differential protection.
Optimized the smoothing factor of a heteroscedastic probabilistic neural network (PNN) using particle swarm optimization.
Extracted the ratio of voltage-to-frequency and differential current amplitude as the primary input features for operating condition classification.
Generated relaying signals across internal faults, external faults, and inrush events via PSCAD/EMTDC transformer simulation and implemented the algorithm in MATLAB.
The optimized heteroscedastic PNN effectively discriminated magnetizing inrush events from internal transformer faults across simulated operating states.
The proposed method showed superior diagnostic discrimination compared to conventional homoscedastic PNN, feedforward backpropagation neural networks, and standard harmonic restraint methods.

Cite This Study

Tripathy et al. (2009) studied this question.

synapsesocial.com/papers/6a7ad49f02ab7a73dd031d26 https://doi.org/10.1109/tpwrd.2009.2028800

View Full Paper

Ask AI

Bookmark

Implication

Simulation study demonstrates accurate discrimination between magnetizing inrush and internal faults in power transformers, indicating improved differential relay reliability.

Power Transformer Differential Protection Based On Optimal Probabilistic Neural Network

Discussion

Member takes

Implication

Key Points

Cite This Study

Power Transformer Differential Protection Based On Optimal Probabilistic Neural Network

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider