Research On Methods Of Flight Accident Event Sequence Estimation Based On Flight Simulation Environment

The characteristics of civil aviation include running environment complicated,high risk,and so on. Once an accident happened, not only cause casualties and economic losses, but also cause huge social and political impact. On the other hand, the causes of the accident analysis is becoming more and more difficult. It is an effective way to realize the reproduction of manipulation state by inverse simulation technology. In numerous inverse simulation algorithms, the integral inverse simulation algorithm is widely used with the advantage of independent with model and simple implementation. In this paper,a simple manipulation of a civil aircraft is reproduced by the integral inverse simulation method based on unconstrained model.The convergence interval and stability are analyzed in the process of integral inverse simulation.The key points of using the inverse simulation algorithm are summarized.Integral inverse simulation method can adapt to unconstrained system well,however,as it doesn’t consider the scope of the state and control input constraints, the result of integral inverse simulation may occur deviation when state variables and control input variables over their range. Therefore, in this paper,the research of inverse simulation with constrained system based on model predictive control is developed.This paper introduced a simple simulation model of the Boeing 737.With its longitudinal model, using the integral method and the model prediction control method to reproduce the manipulation state of elevator rudder surface emergency stop accident.Process of flight accidents usually contain some abnormal sequence of events, this kind of abnormal events can be saw as discrete events. Therefore the nature of the flight accident model is not a continuous system, it is a hybrid system.In this paper,the estimation of flight accident abnormal event sequence for discrete systems with constraint conditions is studied based on hybrid system model.The mixed logical dynamic system model is developed for the scene of one engine failure when aircraft is landing.