Applications of competitive learning and fuzzy associative mapping in classification of power quality disturbances

Proposes a neural-fuzzy technique based classifier for the recognition of power quality disturbances. The classifier adopts neural networks in the architecture of frequency sensitive competitive learning and learning vector quantization. With the neural network, a set of code words is derived from training samples for each type of disturbances. The code words are then exploited to evaluate similarity measures, which in turn are used to determine the disturbance type together with a belief degree through fuzzy associative memory mapping