Proceeding of the Joint International Symposium on Deformatuon Monitoring. 2-4 November, 2011. Hong Kong. China. Report. Dam deformation analysis is one of the main tasks in dam safety monitoring. Regression analysis model is often used in dam deformation analysis in early days. At present, the statistical model, which divides the dam deformation into three parts, hydraulic pressure component, thermal component and ageing component, according to the causes of deformation, has been widely adopted in dam deformation analysis. The BP neural network model and the merging model based on BP neural network algorithm of dam deformation analysis are mainly discussed in this paper, and finally, the four models mentioned above are calculated and analyzed according to a specific project instance. The precisions are respectively ±1.19mm, ±0.38mm, ±0.34mm, and ±0.28mm for single linear regression model, statistical model, BP neural network model and merging model. So it is shown that the merging model is better than the others according to the results.