Evaluation of the multi-year-on-year method for degradation analysis in photovoltaic systems

Photovoltaic (PV) systems are essential for a sustainable energy future, but their long-term performance is affected by degradation phenomena. Accurately quantifying Performance Loss Rate (PLR) is critical for ensuring system longevity, securing financial returns, and optimizing maintenance strategies. Reliable degradation assessments are especially important as PV deployment accelerates globally. Conventional PLR estimation methods, such as the Standard Year-on-Year (YoY) approach, are often subjected to high uncertainties under noisy or incomplete datasets. To address this limitation, the present study investigates whether the Multi-Year-on-Year (Multi-YoY) methodology can provide more robust and reliable degradation assessments under real-world conditions. The Multi-YoY method was implemented on real operational data from eight geographically distributed PV systems located in Denmark, Germany, USA, Poland, Iran, UK, Norway, and Brazil. This approach aggregates overlapping pairs of yearly comparisons, applies Monte Carlo resampling to quantify uncertainty, and refines confidence intervals to obtain more stable degradation estimates, even in the presence of with missing data or high-noise environments. Results show that Multi-YoY consistently outperforms the Standard YoY method, achieving up to tenfold reduction in relative mean error (RME) and significantly narrower confidence intervals. In systems with strong year-to-year variability, the Multi-YoY method provides a clearer and more reliable long-term degradation trend.