Autors: Stoyanov, L. S., Draganovska I. Y.
Title: Comparison of Hybrid Models for PV Power Output Forecasting - Application to Oryahovo, Bulgaria
Keywords: Artificial neural networks;physical model;PV power forecast

Abstract: The paper presents a comparison of different hybrid models used for PV power output forecasting. The models are hybrid, because they use combination of solar radiation forecasting model and power output estimation model. The prediction precision is studied on two time periods - three and six days with different type of solar radiation. The relative root mean square error is used for estimation of the precision. The investigated models show good correlation to the real PV power and can be applied in the exploitation of a PV plant according to the available data.

References

    Issue

    2023 18th Conference on Electrical Machines, Drives and Power Systems, ELMA 2023, 2023, Bulgaria, ISBN 979-835031127-3

    Цитирания (Citation/s):
    1. V. Milenov and Z. Zarkov, "Modeling of Photovoltaic Systems for Self-Consumption," 2023 15th Electrical Engineering Faculty Conference (BulEF), Varna, Bulgaria, 2023, pp. 1-6, doi: 10.1109/BulEF59783.2023.10406197. - 2023 - в издания, индексирани в Scopus и/или Web of Science
    2. Y. Huang, J. Cao and S. Cai, "A Spatial-temporal Neural Network for Photovoltaic Power Prediction," 2023 13th International Conference on Power and Energy Systems (ICPES), Chengdu, China, 2023, pp. 448-452, doi: 10.1109/ICPES59999.2023.10400165. - 2023 - в издания, индексирани в Scopus и/или Web of Science
    3. Evstatiev B., Valov N., Gabrovska-Evstatieva K., Valova I., Kaneva T., Mihailov N., 24 Hours Ahead Forecasting of the Power Consumption in an Industrial Pig Farm Using Deep Learning, 2025, Energies, issue 15, vol. 18, DOI 10.3390/en18154055, eissn 19961073 - 2025 - в издания, индексирани в Scopus
    4. Valova I., Gabrovska-Evstatieva K.G., Kaneva T., Evstatiev B.I., Generation of Realistic Synthetic Load Profile Based on the Markov Chains Theory: Methodology and Case Studies, 2025, Algorithms, issue 5, vol. 18, DOI 10.3390/a18050287, eissn 19994893 - 2025 - в издания, индексирани в Scopus

    Вид: публикация в национален форум с межд. уч., публикация в реферирано издание, индексирана в Scopus