The carbon cycle has a great impact on the environmental changes of the Earth, and fungus play a crucial role in the decomposition of ground litter and woody fibers in nature, so it is very necessary to study the growth process of fungus. First, we used the K-means Clustering Algorithm to classify the different species of fungus into four categories according to their growth rates as well as their moisture tolerance. This not only measures the similarity between various fungus more intuitively, but also reduces the difficulty of subsequent modeling work. Second, we considered environmental accommodation, intra- and interspecific competition, and used this to modify the Lotka-Volterra interspecific competition Model and to derive the relationship between fungal growth rate and time. Then, we used this model for long-term and short-term prediction of fungal growth. In the short term, the growth rate of fungi increased rapidly with time; in the long term, the fungal population tended to be stable.