Photovoltaic power forecasting based on GA improved Bi-LSTM in microgrid without meteorological information

Hao Zhen, Dongxiao Niu, Keke Wang, Yucheng Shi, Zhengsen Ji and Xiaomin Xu

Energy, 2021, vol. 231, issue C

Abstract: Due to flexible and clean nature, distributed photovoltaic (PV) power plants in micro-grid are essential for solving energy and environmental problems. However, because of the high cost of weather station, the meteorological data of distributed power plants is often absent. Therefore, this paper focuses on the accurate output prediction of the target PV station without meteorological data by incorporating the output series of the adjacent PV plants and grasping features by the proposed deep learning models. A novel ultra-short term PV power prediction model based on the improved bidirectional long short-term memory model with genetic algorithm (GA-BiLSTM) is proposed to improve the performance and multiple PV output series are innovatively taken as inputs of the prediction model. A case study is conducted with an actual target PV station in a micro-grid. Sensitivity analysis of input variables is studied and the performance of proposed GA-BiLSTM model is compared with other models under different time horizons to verify the effectiveness. The results illustrate the significance of the output series of adjacent PV plants and the proposed model performs best in the ultra-short term forecasting, with lowest RMSE value of 0.438, 0.806, 1.118 in 5min, 15min, 30min ahead output prediction without meteorological data.

Keywords: Correlated time series; Distributed PV Plants; Deep learning; Power forecast (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (21)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544221011567
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:231:y:2021:i:c:s0360544221011567

DOI: 10.1016/j.energy.2021.120908

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().