Solar insolation forecast using artificial neural network for Malaysian weather

Solar insolation forecast is essential for photovoltaic (PV) generation plant in enhancing the usage of solar energy for electrical production scheme. Likewise, it improves thi PV power generation efficiency by regulating the control algorithm and charge controller corresponding to the prediction pr...

Full description

Saved in:
Bibliographic Details
Main Authors: Chua, H. G., Kok, Boon Ching, Goh, Hui Hwang
Format: Conference or Workshop Item
Published: 2012
Subjects:
Online Access:http://eprints.uthm.edu.my/6074/
http://eprints.uthm.edu.my/6074/1/SOLAR_INSOLATION_FORECAST_USING_ARTIFICIAL.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Solar insolation forecast is essential for photovoltaic (PV) generation plant in enhancing the usage of solar energy for electrical production scheme. Likewise, it improves thi PV power generation efficiency by regulating the control algorithm and charge controller corresponding to the prediction probability. An acquired data in solar insolation is required for solar energy harvesting. This paper presents the 12 hourly solar insolation forecast using Artificial Neural Network (ANN). A Multi-level perceptron (MLP) with back propagation technique model is proposed to predict the next day 12 hours solar insolation. The performance of MLP model is investigated with 60 days of solar insolation data from July 9th to September 9th 2011. The investigation is conducted under two different tropical weather conditions, sunny and rainy conditions. Hence, the best performance MLP forecaster model with a minimal error is selected through a trial and error method under various weather conditions. In this paper, the performance of the forecaster is shown. The results allow inferring the adequate performance and pertinence of this methodology to predict complex phenomena, such as solar insolation.