Classification of Power Signals using ACO based K- Means Algorithm and Fuzzy C-Means Algorithm

This paper presents pattern classification of power signal disturbances using modified form of S-transform, which is obtained by taking the Inverse Fourier transform of STransform is known as time-time transform (TT-transform). The TT-Transform based used for power signals to extract features, visual localization, detection. TT-Transform has good ability in gathering frequency; it gathers the high frequency signals in diagonal position of the spectrum and suppressing the low frequency signals. Only the diagonal of TT-Transform has been used for signal characterization. The diagonal of TT-Transform represent a simple frequency filtered version of the original signal. The extracted features are fed as input to a fuzzy C-means clustering algorithm (FCA) to generate a decision tree. To improve the pattern classification of the fuzzy C-means decision tree, the cluster centers are updated using ant colony optimized technique (ACO). Further K-Means algorithm is used for updation of cluster centers using ant colony optimization technique (ACO) for classification accuracy and the results of both the algorithm are compared.