TITLE:               

Prediction and estimation of PV and WP models based on measurement data

AUTHORS:     

Sukarwoto, Catra Indra Cahyadi, Fauziah Nur, Suwarno

DOI:  10.5110/77. 1134               Page:   280-297          Vol: 19    Issue: 03   Year: 2024

ABSTRACT

An integrated power grid can increase renewable energy sources (RES) in a decentralized power system accompanied by an increase in complex networks. RES integration is very important for estimation and prediction of future network stability and design. However, a single virtual power plant (VPP) is not suitable for large-scale renewable generation. In this paper, a hybrid (PV and WP) computational prediction and estimation model is proposed. Artificial neural networks (ANN) are used for computing. In addition to the predictions and estimates assessed based on the different investigated input features and uncertainties, also the relative mean absolute error (rMAE) on weather conditions such as cold and hot weather on the generating capacity. The results show that there is a difference between the two seasons, where in winter it reaches around 0.48% and in summer around 2.6% for PV power to installed capacity, while PV generation with an average value of around 1.68% and 4.8% for wind generators.

   Keywords:

PV, WP, Integration, Artificial neural networks, generating capacity

Received: 05 March 2024

Accepted: 22 March 2024

Published: 31 March 2024