Risk Early Warning Model for Aircraft Runway Overturn during Thunderstorm Landing Scenarios
Based on the current issues of inadequate quantification in the assessment and early warning of aircraft runway overturns and the neglect of the effects of thunderstorms,a method for aircraft runway overturn risk early warning under thunderstorm scenarios was proposed,integrating a combined weighting method and cloud model.Accident reports were text-mined using the Latent Dirichlet allocation(LDA)topic model to construct a risk early warning index system comprising 4 primary indicators and 24 secondary indica-tors.The G1-CRITIC method was employed to weight the risk indicators,and allocation coefficients were determined through the mini-mum Euclidean distance to calculate combination weights.A comprehensive risk early warning model was established by combining the cloud model with indicator weights.Flight data from both normal and thunderstorm scenarios were selected for risk early warning in-stance verification.The results indicate that under normal scenarios,the comprehensive risk early warning cloud model is most similar to the Level I"Basically Normal"interval standard cloud model,while under thunderstorm scenarios,the comprehensive risk early warning cloud model is most similar to the Level Ⅲ"Moderate Exceeded"interval standard cloud model.This reflects the evolution of aircraft runway overturn risk from normal to thunderstorm scenarios,and the risk early warning results are generally consistent with the actual situation,validating the effectiveness and reliability of the assessment model.This model can provide important references for airport and airline managers in the operation and risk management of flights during thunderstorm weather,as well as in risk mitigation strategies.
aviation transportationrisk early warningaircraft runway overturntext miningcloud modelcombined weighting
万方数据
维普
