The noise level in the aircraft cabin is an important indicator on market competitiveness, and it is a considerable challenge to reduce the noise in the cabin at a relatively small cost, so efficient and accurate forecasting tools are required. In the aviation industry, Statistical Energy Analysis (SEA) is widely used to predict mid- and high-frequency noise in aircraft cabins. It has the advantages of simple method and low requirement of structure details. This paper models the cabin of a certain type of aircraft, and uses statistical energy methods to predict the noise level in the cabin during ground and cruise conditions. Among them, the prediction result of the cabin noise on the ground is in good agreement with the test results above 250Hz, which verifies the correctness of the model. During cruise, the pressure fluctuation on the outer surface of the front fuselage is dominated by turbulent boundary layer noise. The analysis shows that the Robertson model can predict turbulent boundary layer noise well, and the difference between the predicted single-point spectrum and the experimental value is less than 1dB. The cabin noise predicted by only loading the turbulent boundary layer noise source is consistent with the experimental results, but the sound pressure level in some frequency bands is low.