Generalized models for estimation of diffuse solar radiation based on clearness index and sunshine duration in India: Applicability under different climatic zones

Generalized models for assessment of monthly average diffuse solar radiation over India were established using long-term solar radiation data available for 15 years (1986-2000) obtained from Indian Meteorological Department (IMD), Pune. Regression analysis was employed to correlate the diffuse fraction (K̅d)with clearness index (K̅t)and relative sunshine period (S̅S̅o) together. Seven new models (with two input variables i.e. global solar radiation and relative sunshine period) were developed using data of the measurement sites. Well-established models from literature were also compared with the proposed models. Statistical tests used to evaluate the accuracy of models were mean bias error, root mean square error, mean percentage error, coefficient of determination, t-statistics and normalized median absolute deviation. Global performance indicator (GPI) was used to rank the models. Further, the empirical models were applied on the five representative locations under diverse climatic zones (i.e. Hot & Dry, Warm & Humid, Temperate, Cold and Composite climates) prescribed by the Energy Conservation Building Code (ECBC) for India. Proposed models were also compared within each climatic zone and best model was recommended. Developed models were found to have good performance on collective data as well as under each climatic zone individually.