login

Search documents

Browse topics

Document details

 
Title:
 
Fully Automated Photovoltaic System Modelling for Low Cost Energy Management Applications Based on Power Measurement Data
 
Author(s):
 
B. Hanke, M. Bottega, D. Peters, N. Maitanova, J.-S. Telle, M. Grottke, K. von Maydell, C. Agert
 
Keywords:
 
PV System, Modelling / Modeling, Prosumer
 
Topic:
 
PV Systems - Performance, Applications and Integration
Subtopic: Operation, Performance and Maintenance of PV Systems
Event: 35th European Photovoltaic Solar Energy Conference and Exhibition
Session: 6BO.8.2
 
Pages:
 
1588 - 1593
ISBN: 3-936338-50-7
Paper DOI: 10.4229/35thEUPVSEC20182018-6BO.8.2
 
Price:
 
 
0,00 EUR
 
Document(s): paper, presentation
 

Abstract/Summary:


Within this work an automated algorithm for the modelling of photovoltaic systems solely based on smart meter power readings of the system is proposed assuming low cost, stationary rooftop photovoltaic systems. As an example system, data from a nearly horizontal (7° tilt) thin film system is used for method validation. The photovoltaic power output modelling results are compared to an individually engineered INSEL® model and a Linear Regression (LR) Model. The resulting auto model output is discussed in the context of day ahead and intraday photovoltaic power forecast using no-cost / low-cost irradiation information. This work presumes that the characteristics of a photovoltaic system can be extracted from its clear sky /best system power curve. The extracted best system curve can be used directly as a model by scaling the clear sky power output with the ratio of the radiation to the clear sky radiation. The clear sky/best system curves are constructed by identifying the power maximum for a given time of the day for a data set of a sample number of previous days. The auto model is comparable to the fitted INSEL® model and the linear regression (LR). The yearly error metrics for estimation of accuracy of the simulation, concerning 5-minute power values, are between -0.3% for the LR model, 3.7% for the INSEL® model and 4.7% for the auto model. Also the annual RMSE and MAE of the auto model (74.8 W and 27.4 W) are comparable to the LR (66.8 W and 26.2 W) and the INSEL® fit (68.1 W and 25.5 W). The auto model operates with less input data (generated energy only) and no knowledge about the engineering, design or surroundings of the PV system while yielding similar simulation results as conventional methods.