Abstract—Over the past three decades power demand has increased remarkably due to industrialization and increased demand of automation. On the other hand conventional energy sources like fossil fuel are ever depleting. Therefore, industry and scientist are focusing on renewable energy (RE) sources like wind, solar etc., to address twin challenge of energy security and reduction in pollution caused by excessive fossil fuel usage. Further, number of consumers generating renewable energy in distributed manner and participating in the power network is increasing drastically. This exponential rise in penetration of renewable energy into existing power system has posed challenges to grid stability, reliability and power quality. A precise power demand-generation balance is challenging in smooth and reliable operation network, irrespective of unpredictable demand and intermittent nature of renewable power generation. In this research paper we have discussed and designed time series generation forecast models for wind and solar using historical RE generation data for Maharashtra state of India. Forecast results of designed solar and wind power generation models are compared. The wind and solar power generation forecasts obtained in this paper will help the power system operators; while taking decisions related to energy mix, generation planning, scheduling to maintain reliable and economical operation of power system.

Index Terms—Demand supply balance, forecast error, power quality, renewable energy generation forecasting, smart grid, statistical techniques.

Sonali N. Kulkarni is with Research Scholar, Electronics & Telecom Engineering, University of Mumbai, Rajiv Gandhi Institute of Technology, Versova, Andheri (W), Mumbai, Maharashtra, 400053 India. Prashant Shingare is with Director of Renewable Energy, Vertiv Energy Pvt. Ltd, NITCO Business Park, Wagle Industrial Estate, Thane (W), Maharashtra, 400604 India.

Cite:Sonali N. Kulkarni, Prashant Shingare, "Generation Forecasting Models for Wind and Solar Power," International Journal of Computer and Electrical Engineering vol. 10, no. 4, pp. 318-329, 2018.