Advancements in PV Plant Energy Production Prediction with Model Improvement Based on Measured Data

摘要

The object of this research is to implement modifications into conventional analytical model for calculating energy production from a PV plant. Improvements are introduced through equipment work characteristics and Liu-Jordan-Klein model modifications used in conventional analytical model predictions for the PV plant in Varazdin, Croatia. A two-year measured energy production data is available and is stored into a database. The original conventional analytical model selected for this research is presented with the Homer software tool. The improvements are based on one year period measured data to get more accurate results on hourly basis, and therefore ensure better results for monthly and annual values. The second year period measured data is used to verify the introduced modifications. It can be expected that verification of the model will generate more inaccurate results, especially in transition periods. If model output results for the second year are more accurate than the original conventional model, it can be concluded that the improvements are rather correct. It is estimated that at least 5 years of measuring process are needed in order to generate the best set of correction factors. Methods for implementation of improvements are now developed for the available data and results, and are presented in this paper.