Analysis of Various Machine Learning Algorithms on Wind Power Forecasting with Introduction of Varying Rates of Missing Data

Authors

  • Rakshith Dasenahalli Lingaraju Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, USA

Keywords:

Machine Learning, K Nearest Neighbor, Support Vector Regression, Mean Square Error, Wind power forecasting, Randomized missing data

Abstract

Machine learning (ML) is a process which reads through data and finds statistical relationships and structure within the data which can be used to make predictions. Various industry sectors have implemented ML and are reaping great benefits. However, the energy sector has been slow in its adoption and has yet to see its full potential. This paper aims to address how Machine Learning can greatly benefit the energy sector in optimizing and improving accuracy of wind power forecasting. This paper compares performance of two algorithms: Support Vector Regression (SVR) [1] and K-Nearest Neighbor (KNN) [2] in wind power forecasting and will be evaluated based on the Mean Square Error (MSE). The true predictive capability of the trained model will be analyzed when data points at random are removed to test their effect on MSE of the algorithms.

Downloads

Download data is not yet available.

Downloads

Published

01-12-2022

Issue

Section

Articles

How to Cite

[1]
R. D. Lingaraju, “Analysis of Various Machine Learning Algorithms on Wind Power Forecasting with Introduction of Varying Rates of Missing Data”, IJRESM, vol. 5, no. 11, pp. 213–217, Dec. 2022, Accessed: Jun. 17, 2024. [Online]. Available: https://journal.ijresm.com/index.php/ijresm/article/view/2450