Performance Characterisation of Utility-Scale Photovoltaic Systems Operating in Semi-Arid Conditions

Abstract

Aims: The study investigates the operational performance of a utility-scale grid-connected photovoltaic (PV) system operating under hot semi-arid conditions of Mali. Study Design: The analysis was based on key performance indicators: system efficiency, performance ratio (PR), capacity factor (CF), inverter efficiency, energy yields, and cell temperature, to assess the influence of climatic conditions on energy production and conversion efficiency. Place and Duration of Study: The study was conducted on a utility-scale grid-connected PV system in Mali. The analysis was based on data collected over a one-year operational period. Methodology: The system’s operational performance was evaluated by analysis key performance indicators: system efficiency, performance ratio (PR), capacity factor (CF), inverter efficiency, energy yields, and cell temperature. Statistical methods, including regression analysis, were used to determine the relationships between these parameters and climatic variables. Results: Results show that the system maintained a stable efficiency of 8.8 ± 0.2%, while inverter efficiency remained consistently high at 95.5%, corresponding to annual inverter losses of only 4.5%. The annual average PR and CF were 73.5% and 17.8%, respectively, equivalent to approximately 1,560 full-load hours. A strong inverse relationship was observed between PR and cell temperature (R² = 0.78), with a sensitivity of -0.24 percentage points per °C, indicating that thermal effects are the main source of seasonal performance variations. In contrast, the relationship between PR and DC output was weak (R² = 0.11), demonstrating that fluctuations in energy production are primarily driven by solar resource availability rather than degradation or conversion inefficiencies. A strong linear relationship was found between CF and DC output (R² = 0.94,p < 0.001), confirming the predictability of energy production. Yield analysis further revealed highly proportional relationships between reference, array, and final yields, with coefficients of determination exceeding 0.93 and a mean \frac{Y_{f}}{Y_{r}} ratio of 0.735, identical to the average PR. Clipping losses were limited to 1.44%, despite a DC/AC ratio of 1.31, indicating appropriate system sizing. Conclusion: Overall, the results demonstrate that the PV system exhibits high technical reliability, limited losses, and strong resilience to the climatic conditions of the Sudano-Sahelian region, supporting the long-term viability of utility-scale solar deployment in semi-arid environments.