oalogo2  

AUTHOR(S):

Jyotsana Pandey, Aaditya Khare

 

TITLE

Implementation of DWT and Regression Learning Based ANN for Forecast of Solar PV Output

pdf PDF

ABSTRACT

The intermittent nature of PV generation makes it difficult to plan load dispatch from grids which have PV integration. One of the most effective ways to predict PV power is to estimate the solar irradiation on the PV cells which typically vary to month, day, time, temperature and other variables. The sporadic nature of the irradiation on PV cells males it extremely challenging to prediction of short term and long-term PV outputs. The proposed work presents a DWT-regression learning based neuro network predictor for solar PV generation output for both short- and long-term conditions. The performance of the system has been evaluated based on the mean absolute error, number of iterations and the regression curves. It can be observed from the results that the proposed system attains extremely low mean absolute error for the PV output thereby rendering high accuracy to the micro/smart grids with PV integration

KEYWORDS

Photovoltaic (PV) output Prediction, Neuro-Network Predictor, Mean Absolute Error, and Regression Learning

 

Cite this paper

Jyotsana Pandey, Aaditya Khare. (2021) Implementation of DWT and Regression Learning Based ANN for Forecast of Solar PV Output. International Journal of Circuits and Electronics, 6, 17-22

 

cc.png
Copyright © 2021 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0