To address the issues of low denoising accuracy and the misclassification of useful high-frequency information as noise in existing methods for dam deformation monitoring data, a denoising method combining the adaptive noise complete ensemble empirical mode decomposition (CEEMDAN) and improved wavelet threshold was proposed. This method decomposes the original data using CEEMDAN and performs feature analysis on the intrinsic mode function (IMF) components obtained from the decomposition through t -tests, screening out noisy components. These components were verified using the Pearson correlation coefficient and variance contribution rate. Finally, the identified noise-containing components are finely denoised using an improved wavelet thresholding method, and the denoised IMF components are reconstructed to obtain the denoised data. Simulation tests and engineering case verification results show that this method outperforms three comparative methods across various indicators, while effectively preserving useful high-frequency information, improving accuracy and smoothness, and can be used for denoising nonlinear deformation data in dams.