Many important lakes and reservoirs of China, including Lake Taihu, Lake Chaohu, Lake Dianchi, Lake Erhai and Three Gorges Reservoir, were plagued with cyanobacterial blooms. However, the intensity of the blooms in these freshwaters varied significantly in different years, which exhibited significant challenges to the blooms collection organizations and drinking water plants, leading to the urgent need to cyanobacteria blooms prediction model based on annual dataset. Therefore, the long-term (15 years) observation data and meteorological and hydrological datasets of Lake Taihu were collected for the prediction of algal blooms. In current study, cyanobacterial bloom intensity index (BI) were proposed with the consideration of yearly average blooms area interpret by high frequency remote sensing images and whole lake average chlorophyll-a concentration. Furthermore, environmental factors, such as water temperature, rainfall, water level, nitrogen and phosphorus concentrations were used as the crucial factors to predict BI. Our results showed that average water temperature in winter and early spring, as well as the rainfall of the former year were significant positive factors of the yearly BI value in Lake Taihu. While the nutrient-related factors in early spring had no significant relationships with BI. In addition, a multiple (or univariate) regression analysis based on the above factors (BI was the dependent variable and the remaining environmental factors were the independent variables) were performed in this study, and the optimal model was selected. In general, the predicted results of the selected optimal model had a high consistency with the measured concentrations, thus the model obtained in this study had relatively high accuracy for predicting the interannual intensity of cyanobacteria blooms in Taihu Lake. This study may serve reliably for the medium- and long-term prediction of cyanobacteria blooms in Lake Taihu, and other eutrophic lakes.