Authors: Ranjeeta Bisoi, P.K. Dash
Addresses: Siksha 'O' Anusandhan University, Bhubaneswar-751030, Orissa, India. ' Siksha 'O' Anusandhan University, Bhubaneswar-751030, Orissa, India
Abstract: This paper presents a new approach for power signal time series data mining using S-transform (ST) based K-means clustering technique. Initially the power signal time series disturbance data are pre-processed through an advanced signal processing tool such as ST and various statistical features are extracted, which are used as inputs to the K-means algorithm for disturbance event detection. Particle swarm optimisation (PSO) technique is used to optimise cluster centres which can be inputs to a decision tree for pattern classification. The proposed hybrid PSO-K-means clustering technique provides accurate classification rates even under noisy conditions compared to the existing techniques, which shows the efficacy and robustness of the proposed algorithm for time varying database like the power signal data.
Keywords: K-means clustering; time-varying power signals; power signal data; particle swarm optimisation; PSO; S-transform; decision tree; classification; data clustering.
International Journal of Data Mining, Modelling and Management, 2011 Vol.3 No.3, pp.277 - 302
Published online: 06 Aug 2011 *Full-text access for editors Access for subscribers Purchase this article Comment on this article