An expert system based on S-transform and neural network for automatic classification of power quality disturbances

摘要: In this paper, an S-transform-based neural network structure is presented for automatic classification of power quality disturbances. The S-transform (ST) technique integrated with (NN) model multi-layer perceptron to construct the classifier. Firstly, performance ST shown detecting and localizing disturbances by visual inspection. Then, used extract significant features distorted signal. addition, optimum combination most useful identified increasing accuracy classification. Features extracted using are applied as input NN (PQ) that solves a relatively complex problem. Six single two well pure sine (normal) selected reference considered Sensitivity proposed expert system under different noise conditions investigated. analysis results show classifier can effectively classify PQ