Landing gear system health assessment based on SVM and XGBoost dual-flow modeling

The landing gear system stands as one of an airplane's most crucial components, playing a pivotal role during take-off and landing phases. Its failure could result in catastrophic accidents, including belly landings, with severe consequences such as loss of life and aircraft destruction. Comprising a complex integration of mechanical, electrical, and hydraulic components, the landing gear system operates within a non-pneumatic zone, subjected to harsh working environments. Potential safety hazards, including mechanical wear, inadequate lubrication, and hydraulic equipment leakage, can compromise the system's extension capability, and in severe cases, lead to extension or retraction failure. Hence, effective management of landing gear health and early detection of system insecurities are imperative for ensuring aircraft safety. In this paper, a dual-flow model was developed based on SVM and XGBoost, facilitating accurate prediction of landing gear retraction and release times under varying external environmental conditions. This model serves as a foundation for assessing the health state of landing gear systems.