Wind Speed Forecasting

Abstract

Abstract As the world exhausts its non-renewable energy reservoirs, building predictive models for renewable energy dependencies comes important a fortiori. In this study we examine how well a Deep Neutral Networks performs on wind speed data in a time series forecasting. The data used are based on wind speed readings acquired at the first-of-its-kind LiDAR based offshore which is situated at the Gulf of Khambhat, Gujarat, which is about 23 km from the Gujarat Coast. The performance has been calculated on the basis of average absolute error and average squared error. The period of data taken for analysis is from 1st of November 2017 to 30th November 2018, separated at 10-minute intervals. The data set provides wind speed data from 40 m to 200 m. We are taking the 100 m and 200 m cases for our analysis. The predictions made can further be used to calculate the power generation of the wind power plant. Keywords: Wind Speed Forecasting, DNN, Standard Scalar, Matplotlib Cite this Article Neeraj Kumar, Sumit Mishra, Ashutosh Dubey, Abhishek Dhiman, Tanmay Baweja. Wind Speed Forecasting. Recent Trends in Electronics & Communication Systems . 2020; 7(2): 13–17p.