The invention discloses a low earth orbit satellite mobile communication resource intelligent scheduling method. The method comprises the following steps: S1, initializing deep learning network parameters; s2, inputting prior low-orbit satellite sample data, and fitting the sample data by using a deep learning network training method; s3, analyzing the sample deviation, if the fitting deviation isgreater than 0.01, indicating that the existing deep learning network belongs to an under-fitting state, and turning to the step S2; otherwise, entering the step S4; s4, low-orbit satellite sample data is collected again, and state marking is carried out to serve as a test sample; s5, analyzing the variance of the sample, if the fitting variance is greater than 0.005, indicating that the currentsample is in an over-fitting state, performing regularization processing on the deep learning network, and turning to the step S2 after the processing is completed; otherwise, entering the step S6; s6, sample data needing to be predicted of the low-orbit satellite is input, and a low-orbit satellite distribution network resource strategy is obtained. And intelligent scheduling of low-orbit satellite mobile communication resources can be better realized.