During the operation of a photovoltaic inverter, an increase in temperature significantly affects the output power, efficiency, service life, and stability of the photovoltaic inverter. However, traditional evaluation methods based on physical experiments are time-consuming and costly, making it difficult to fully capture the dynamic performance under real conditions. This paper proposes a DT-LSTM (Digital Twin-Long Short-Term Memory) output power prediction method based on the thermoelectric coupling digital twin model of the inverter. Firstly, a digital twin model based on the equivalent circuit of the inverter is established; secondly, internal physical field data are calculated and parameter identification is performed; finally, the inverter output power is predicted and compared with traditional methods. The research results show that the prediction accuracy of the proposed method is improved, with the Mean Absolute Percentage Error (MAPE) optimized to 7.81%.
Marques LameirinhasRATorresJPNde Melo CunhaJP. A photovoltaic technology review: history, fundamentals and applications. Energies2022; 15(5): 1823.
2.
VoTTEKoHHuhJ, et al.Overview of possibilities of solar floating photovoltaic systems in the offshore industry. Energies2021; 14(21): 6988.
3.
GaboitaolelweJZungeruAMYahyaA, et al.Machine learning based solar photovoltaic power forecasting: a review and comparison. IEEE Access2023; 11: 40820–40845.
4.
SoomarAMHakeemAMessaoudiM, et al.Solar photovoltaic energy optimization and challenges. Front Energy Res2022; 10: 879985.
5.
HøiaasIGrujicKImenesAG, et al.Inspection and condition monitoring of large-scale photovoltaic power plants: a review of imaging technologies. Renew Sustain Energy Rev2022; 161: 112353.
6.
HazimHIBaharinKAGanCK, et al.Review on optimization techniques of PV/inverter ratio for grid-Tie PV systems. Appl Sci2023; 13(5): 3155.
7.
BoscainoVDittaVMarsalaG, et al.Grid-connected photovoltaic inverters: grid codes, topologies and control techniques. Renew Sustain Energy Rev2024; 189: 113903.
8.
HazimHIAzmi BaharinKKim GanC, et al.Techno-economic optimization of photovoltaic (PV)-Inverter power sizing ratio for grid-connected PV systems. Results Eng2024; 23: 102580.
9.
ZhangBGaoY. IGBT reliability analysis of photovoltaic inverter with reactive power output capability. Microelectron Reliab2023; 147: 115073.
10.
ZhaoESHanYLinXY, et al.Harmonic characteristics and control strategies of grid-connected photovoltaic inverters under weak grid conditions. Int J Electr Power Energy Syst2022; 142: 108280.
11.
MalikAHaqueABharathKVS, et al.Overview of fault detection approaches for grid connected photovoltaic inverters. e-Prime-Advances in electrical engineering. Electron Eng2022; 2: 100035.
12.
TaoFXiaoBQiQL, et al.Digital twin modeling. J Manuf Syst2022; 64: 372–389.
13.
LiuXJiangDTaoB, et al.A systematic review of digital twin about physical entities, virtual models, twin data, and applications. Adv Eng Inform2023; 55: 101876.
14.
WangYGQiYLiJ, et al.The wind and photovoltaic power forecasting method based on digital twins. Appl Sci2023; 13(14): 8374.
15.
NatgunanathanIMak-HauVRajasegararS, et al.Deakin microgrid digital twin and analysis of AI models for power generation prediction. Energy Convers Manag X2023; 18: 100370.
16.
ZazoumB. Solar photovoltaic power prediction using different machine learning methods. Energy Rep2022; 8: 19–25.
17.
MarkovicsDMartinJM. Comparison of machine learning methods for photovoltaic power forecasting based on numerical weather prediction. Renew Sustain Energy Rev2022; 161: 112364.
18.
ZhangXYLiYSLiTY, et al.Digital twin empowered PV power prediction. J Mod Power Syst Clean Energy2023; 12(5): 1472–1483.
19.
AlghamdiHMaduabuchiCAlbakerA, et al.A prediction model for the performance of solar photovoltaic-thermoelectric systems utilizing various semiconductors via optimal surrogate machine learning methods. Eng Sci Technol Int J2023; 40: 101363.
20.
ZayedMEAboelmaarefMMChazyM. Design of solar air conditioning system integrated with photovoltaic panels and thermoelectric coolers: experimental analysis and machine learning modeling by random vector functional link coupled with white whale optimization. Therm Sci Eng Prog2023; 44: 102051.
21.
Al-SelwiSMHassanMFAbdulkadirSJ, et al.RNN-LSTM: from applications to modeling techniques and beyond—Systematic review. J King Saud Univ Comput Inf Sci2024; 36(5): 102068.
22.
JebliIBelouadhaF-ZKabbajMI, et al.Prediction of solar energy guided by pearson correlation using machine learning. Energy2021; 224: 120109.
23.
NegroMRobertoG. Advanced 3D electromagnetic transient-thermal steady coupled FEM simulation of a LV busbar of a rectifier transformer. Int J Appl Electromagn Mech2024; 76(1-2): 257–269.
24.
KonstantinouMPeratikouSCharalambidesAG. Solar photovoltaic forecasting of power output using lstm networks. Atmosphere2021; 12(1): 124.
25.
AkhterMNMekhilefMSMokhlisH, et al.A hybrid deep learning method for an hour ahead power output forecasting of three different photovoltaic systems. Appl Energy2022; 307: 118185.
