Processed PV Degradation Dataset for Hybrid-Censored Lifetime Modeling (PVDAQ Systems 10000 and 10003)

This dataset contains processed photovoltaic (PV) system performance data used in the study of degradation under imperfect monitoring conditions. The data are derived from the NREL PVDAQ database and the NSRDB meteorological database for two utility-scale PV systems (System 10000 and System 10003). The dataset includes daily observations of AC energy output, normalized energy performance, ambient temperature, and irradiance variables after quality control and preprocessing. Data cleaning procedures include filtering invalid observations, handling missing values, and normalization of daily AC energy relative to a system-specific baseline. In addition, the dataset incorporates stage-based discretization (k = 7) and flags indicating whether observations were used in model fitting or trajectory-based evaluation. These processed data support hybrid-censored lifetime modeling of degradation and the estimation of time-to-threshold performance metrics such as T90. This dataset is intended to support reproducibility of the results presented in the associated study on hybrid-censored generalized Lindley modeling with robust estimation using density power divergence. Data sources:- NREL PVDAQ (Photovoltaic Data Acquisition System)- NSRDB (National Solar Radiation Database) The dataset includes the following files:- processed_system10000.csv- processed_system10003.csv- README file with detailed variable definitions This dataset provides a reproducible foundation for lifetime-based degradation analysis of photovoltaic systems under incomplete and noisy monitoring conditions.