The invention belongs to the technical field of engine modeling and provides a method for modeling the starting process of a micro gas turbine engine. The method comprises the following steps: modeling the rotational speed of the micro gas turbine engine; modeling the relationship between performance parameters and speed of micro gas turbine engines; performing error analysis. Existing modeling methods of micro gas turbine engine are mostly based on pure mechanism or machine learning, which is difficult to accurately describe the starting process of micro gas turbine engine, and machine learning requires a large number of experimental data. In the engineering practice, the pure mechanism operation is complicated, the efficiency is low, and the modeling accuracy is not high. The invention provides a method for modeling a starting process of a micro gas turbine engine based on the combination of a mechanism and an identification method, which makes up for the shortcomings of the prior art. The invention has the advantages of simple operation and high accuracy, and can realize the modeling of the whole micro gas turbine engine. This method has certain expansibility and can be extendedto other fields.