Short term wind power forecasting using k-nearest neighbour (KNN). Issue 1 (2nd January 2022)

• Record Type: Journal Article
• Authors: Mahaseth, Rahul; Kumar, Neeraj; Aggarwal, Aayush; Tayal, Anshul; Kumar, Amit; Gupta, Rajat
• Is Part Of: Journal of information & optimization sciences. Volume 43:Issue 1(2022)
• Page Start: 251
• Page End: 259
• Publication Date: 2022-01-02
• Subjects: 68xx; Renewable energy -- Machine learning -- K-nearest neighbour -- Decision tree -- Least absolute shrinkage selector operator -- Xtreme gradient boost; Electronic data processing -- Periodicals; Information science -- Periodicals; Mathematical optimization -- Periodicals; 519.6
• DOI: 10.1080/02522667.2022.2042093
• Languages: English
• ISSNs: 0252-2667

Abstract

Abstract: This project focuses on prediction of energy power based on data of previous 2 years using various machine learning algorithms. The data is analysed on yearly basis. The wind power analysed is of four different locations. After the selection of location their given data was tested with various machine learning algorithms and were applied to the dataset and different wind power generation value was predicted depending on location, geographical, demographic and wind speed and weather conditions. On working with different regression techniques, we found out that KNN Algorithm was found to be most effective for the prediction of wind power. Hence, various machine learning and deep learning techniques were applied to get an accurate idea regarding the establishment of wind power plants in particular location analogous to the places selected in the dataset.