Wind power is a prominent and viable alternative and to traditional sources. In India, Ministry of New and Renewable Energy (MNRE) has already estimated 103 GW of wind energy potential. Government of India is looking at renewable energy as a way to achieve energy security and to curb greenhouse gas (GHG) emissions happening due to the electricity sector. Therefore, renewable energy development is high on government agenda. Recently, India set 60 GW of wind capacity additional target by 2022. So the Wind frame performance evaluation is one of the most essential and critical activities in wind plant. The wind frame performance will cover both technical and economical aspects. Woon L W etc al (2014), discussed various machine learning techniques for supporting Data Analytics for Renewable Energy Integration. Pimpalapure P and Jain S (2016), worked out three different turbines cut in speed through machine learning techniques with help of National Renewable Energy Laboratory data. They developed power curve model speed and power using Gaussian Regression.
In previous literature, frame performance depends on wind speed, wind direction, rotor diameter, hub height etc.,
This Research first part proposes Autoregressive Integrated Moving Average (ARIMA) processes to select the appropriate stochastic model for wind speed forecasting in wind frame in India and develops empirical study of modeling and forecasting time series data of wind speed in India.
This Research second part proposes and demonstrates the application of Machine Learning Techniques in evaluating and predicting the performance measures of wind frame in India.
The result of the research can be used to identify best wind frame for the purpose of maximizing capacity, efficiency, life time and minimizing investment and operating cost.