Determination of temperature coefficients tuning for PV modules from empirical data

The photovoltaic literature covers a wide range of translation methods for the I–V curves translation to other conditions of irradiance and cell temperature. The traditional methods have certain limitations, especially in the presence of aging or degradation effects. In this work, a new determination method of temperature coefficients tuning for PV modules from empirical data is demonstrated. Therefore, a dataset of empirical outdoor I–V measurements collected under varying temperature and irradiance conditions. To evaluate the performance, the proposed method treats the Isc and VOC temperature coefficients (α and β) as dynamic, condition-dependent quantities estimated through pairwise transitions between operating states (Ti, Gi) → (Tj, Gj). A matrix (αi,j, βi,j) is thus constructed. When applied to 88 distinct operating conditions, the framework yields modal coefficients of α = −0.0005 mA·°C−1 and β = −86.96 mV °C−1. Validation demonstrates excellent agreement between measured and reconstructed I–V and P–V curves, with correlation coefficients exceeding 0.999 and a maximum power point error of less than 0.32 W, with relative error below 0.42%. While, current NRMSD does not exceed 3% for any case. The proposed method’s novelty lies in three main contributions: (i) A new temperature coefficients adjustment method for the I–V correction procedure is proposed for degraded PV modules, using a mathematical model; (ii) The method demonstrates resilience to outdoor measurement conditions; (iii) a fully empirical outdoor calibration procedure applicable to PV modeling and adaptive performance monitoring. Challenges and directions for future work are also presented.