The T₂ spectrum inversion from NMR data has been widely used in petroleum industry both in core analysis for petrophysical studies and in logging interpretation. In order to improve and consummate algorism of inversion, the authors used the NMR numerical simulation data to study the factors influencing T₂ spectrum inversion from NMR data, such as the intervals and number of echos, and signal-to-noise ratios. The method of T₂ spectrum inversion was compared with optimization algorithm and singular value decomposition (SVD) at different signal-to-noise ratios at the same time. The result shows that TE should match the number of echos and components to attain the best inverse effect when the NMR relaxation signal is collected. Higher signal-to-noise ratio can reach better effect, and when the signal-to-noise ratio is larger than 25, the inverted spectrum coincides with designed spectrum. The method of T₂ spectrum inversion with optimization algorithm doesn't require high signa-to-noise ratio, and the method of SVD could clip inversion.