Evaporation is an important component of the distribution of energy and the hydrological cycle on the global scale, and is also an indicator of climatic, environmental, and ecosystem changes. In this paper, eddy covariance data from Dapukou station in 2012 are used to analyze the seasonal and diurnal variation of evaporation across Lake Taihu. In addition, 11 evaporation models are also evaluated. The results show that the total evaporation from Lake Taihu in 2012 was 1066.2 mm. The latent heat flux was the dominant term of the net radiation energy, which occupied about 91.9% in 2012. The water body stores heat in the period from February to July, after which the net radiation increases to a maximum in July, and evaporation also reaches a maximum. Thereafter, net radiation reduces to a minimum in December, stored heat is released, and evaporation reaches a minimum in February. Comparing 11 evaporation models with the eddy covariance data to evaluate their performance with respect to the evaporation from Lake Taihu, we discuss the total annual evaporation and evaporation variation. The Bowen ratio energy budget model shows the best performance, because the correlation coefficient between the simulation value and the observation is 0.99, and the centered-mean-square error is 4.50 mm month^-1.