Abstract Harvesting energy from renewable sources has become prominent since the use of fossil fuels became unsustainable. Traditional practice for mitigating the energy demand around globe majorly consists of utilizing conventional sources and injection of renewables as and when available. The continuous and exponential growth in consumption alongside the need to reduce the carbon footprint and to counter the climate change has paved the way for Renewable Energy Sources (RES). Availability and maturity in technology made wind and PV (photo-voltaic) the most prominent among others. Per contra, the inherent variations in the weather in form of wind speed, solar irradiance act as a barrier in utilizing the full potential. The variations, ramp events, in case of wind energy have adverse effects on determining the reliability, economical profitability, and flexibility. Accurate recognition of the wind ramp events can improve energy management, forecasting and causality. This paper proposes a data analysis oriented approach exploring the pre-processing technique of wind power variations using moving average filter, followed by noise extraction and separating the power swings. Further clustering the power swings utilizing K-means clustering technique. The proposed technique improves the power swings identification process by reducing the noise content.