DNI assessment is essential for developing any CSP projects. In this manuscript, an approach is made for the analysis and estimation of DNI by developing the empirical models for India. The historical data of around 15 years (1986-2000) of DNI and cloud cover (CC) from six representative locations of major climatic zones namely Bhopal, Jaipur, Srinagar, Patna, New Delhi & Thiruvanantpuram have been employed. From these locations, first four have been used to develop the models correlating DNI and cloud cover
however, rest two were used to evaluate their validity. The standard statistical predictor's viz. CoD, MPE, MBE, RMSE, skewness, kurtosis & chi-square were applied to evaluate these empirical models. The CoD was observed greater than 0.8 while performance criteria have low and acceptable values (MPE-11.4% to -13.77%
MBE 0.49 to 0.96
RMSE 1.22 to 1.44
Skewness 0.90 to -0.15
Kurtosis 0.90 to -1.36 and Chi-square 6.87 to 8.50). The adequecy of empirical models was evaluated through comparing estimated and measured DNI for New Delhi and Thiruvanantpuram which coincides very closely. The study established an approach to estimate DNI using cloud cover data which may facilities CSP projects developers for site selection and decision making in the absence of ground data.