Abstract: Taking the Cunninghamia lanceolata plantation in Yangkou State-owned Forest Farm, Shunchang, Fujian Province, as the research object, 6 samples of C. lanceolata plantations of different age grades were set up. The diameter, DBH and tree height of C. lanceolata trunk at different relative ground heights were extracted by terrestrial laser scanning(TLS) and point cloud processing technology. The age grade was introduced as a dummy variable to construct the C. lanceolata dummy variable taper equation model, screening and testing of model fitting results through parameter t-test, decisive coefficient, and estimation accuracy. The results show that the fitting evaluation index of the model 4 variable parameter taper equation is the best. The result shows that among 6 alternative taper equations, the fitting evaluation index of model F4 variable parameter taper equation is the best, which can be used as the basic taper equation. Through t-test screening of different dummy variable parameter combinations, it was found that when the dummy variable combination Z7 parameters were c and i, each fitting evaluation index was the best, its R² was 0.98 higher than other combinations, and the evaluation index was the best, which could be used for the construction of dummy variable model. Model constructed with dummy variable combination Z7, the model decision coefficient R² is 0.96, the prediction accuracy P is 99.63, the goodness of fit and the prediction accuracy is better than the basic taper equation, the discrete degree of the dummy variable model is significantly smaller, R² is increased by 14.2%, and the prediction accuracy P is increased by 0.55%. Based on TLS technology, constructing a model with age class type dummy variables has better prediction accuracy than the basic taper equation model, which can effectively predict the diameter of any node in the trunk shape of C. lanceolata and simulate the trunk shape of C. lanceolata.