Performance Analysis of a Solar Farm Capturing a Unique Real-time Performance Ratio through Data-driven Methodology

Abstract

Photovoltaic technology, a rising renewable energy source, relies heavily on irradiation and temperature for performance. Defects like shading, hot spots, and soiling can disrupt the current–voltage curve, challenging inverters to operate at maximum power point (MPP). Indirect measurement and real-time loss detection methods are essential due to the inaccessibility of PV cell circuits. The conventional methods of performance measurement consider the performance based on standard test conditions. The main focus of the present work is to develop a novel performance ratio, which captures real-time efficiency involved in Solar farm operations using data-driven methodologies. The major contributions of this work are: Monitoring health and performance of solar farms by collecting 8 months of SCADA data using IEC standard 61724; Developing a data-driven model to predict ideal I-V curve and MPP at each transient state in the absence of panel-level data; Developing a new performance ratio which provides insights into the transient operational state of the inverter and the deviation of real-time data from the MPP. The results reveal the interplay of different inverters and the evolving overall performance of solar farm over time. This underscores necessity for maintenance and highlights the potential for enhancing solar farm’s output.