Machine learning-based low earth orbit satellite orbit forecast precision improvement model establishment method

The invention relates to a method for establishing a low-orbit satellite orbit forecast precision improvement model based on machine learning. Comprising the following steps: generating orbit truth value data XTrue under a full dynamic model, and orbit estimation data XEst and orbit prediction data XPre under a preset dynamic model by adopting precise numerical value extrapolation software; obtaining a track truth value error according to the XTrue and the XPre, and obtaining a track relative forecast error according to the XEst and the XPre; based on an XGBoost model, determining a preset input characteristic variable by taking the track true value error as a target variable, and performing normalization processing; analyzing the normalized preset input characteristic variable and the target variable by using an XGBoost model, and selecting a preset input characteristic variable combination with the maximum determination coefficient R2 as a key input characteristic variable; performing hyper-parameter optimi