Global solar radiation (G) and ultraviolet radiation (UV) observed at Wuhan, China from 2006 to 2012 have been used to investigate the temporal variability of both radiant fluxes and the UV/G ratios (FUV) at different time scales for the first time in central China. Clearness index (Kt) was used to study the cloud effects on FUV in each month under different sky conditions. It turned out that FUV reached higher values in summer and lower values in winter; FUV also increased generally with cloudiness. A UV model for cloud-free condition was developed by studying the dependence of hourly UV irradiations on the relative optical air mass (m), which has been assessed through the statistical indices: mean bias error (mbe), mean-absolute bias error (mabe) and root-mean-square error (rmse) whose values were 0.32, 7.3 and 8.64%, respectively. UV clearness index, m and Kt were further used for analysing the cloud effects, and different UV models under any sky conditions have been established and validated. Meanwhile, as a modulation for different sky conditions, both cloud modification factors were also used in the construction of a new type of UV model. Finally, by comparing the statistical indices from different models, the most suitable model was chosen for validating at Wuhan University (WHU) and two other sites in different regions of China under any atmospheric conditions, which suggested that the proposed model should be modified to account for local differences to produce better estimations in larger areas.