Short-term solar power forecasting- An approach using JAYA based recurrent network model.

Autor: Gundu, Venkateswarlu, Simon, Sishaj P., Kumba, Krishna
Zdroj: Multimedia Tools & Applications; Mar2024, Vol. 83 Issue 11, p32411-32422, 12p
Abstrakt: Solar power generation is variable and weather-dependent, making accurate forecasting essential for power system operation and planning. In this work, we use multiple deep neural network models and different weather parameters to forecast solar power generation. We compare the performance of the distinct models and test the anticipated forecast model using 100 kW solar data from a specific region in Asia in the year 2021. Here, the proposed work uses a deep recurrent network model that is Long Short Memory Network (LSMN) for solar power forecasting. In this work, we used the JAYA algorithm for hyperparameters optimization to demonstrate their forecast strength related to standard analysis. The forecasted results using JAYA-based LSMN show superior predicting performance compared to the conventional techniques. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index
Externí odkaz: