Flight Safety Detection In Landing Phase Based On Gaussian Mixture Model

Flight safety plays an important role in the development of aviation industry.The key of safety management is the effective control of safety risks,and the necessary prerequisite for control is to determine the possibility that causes safety risks.The purpose of flight safety detection is to determine the possibility of safety risks involved in commercial flights,and make the correct identification of whether the flight is safe or not.After that,the possibility of causing safety risks that are not accepted by the system can be control.Although current civil aviation safety management system has been able to effectively identify the safety risks arising from the operation of aircraft,there are still some abnormal operation hidden in the large amount of non-exceeding data generated by the daily operation of flights,and the possibility of risks contained in these abnormal operation is often not accepted by the system.By detecting these abnormal operation that are more likely to cause safety risks,the safety of commercial flights can be detected.While the safety of flights is divided,the value of a large amount of non-exceeding flight data in operational risk management can be used.And the research can also provides the basis for abnormal operation’s analysis and control measures.In order to detect flight safety of commercial flights during landing phase,a flight safety detection method for landing phase is proposed based on the Gaussian Mixture Model.Firstly,by combining Bayesian Information Criterion and K-Means algorithm,the non-exceeding data of approach and landing phase was submitted to the Expectation-Maximization algorithm to construct the Gaussian Mixture Model.Then the temporal distribution of each Gaussian component was studied to characterize each Gaussian component’s appropriateness with time.And the calculation results of the probability density functions of Gaussian components were obtained.After that,a function was difined to judge whether a flight’s data was normal or not at a certain moment.And by comparing the function value of every second with the preset detection threshold,it was able to identify flights with abnormal flight data during approach and landing phase.Finally,experts in the field of flight safety would review the original flight data of abnormal flights to detect and analyze abnormal operation.This method is used to detect flight safety of 462 commercial flights from a specific airline,and by comparing the decision value of each flight with the three different detection thresholds,27,15 and 8 flights with abnormal data are identified respectively.After that,3 experts in the field of flight safety are invited to reviewed 15 flights with abnormal flight data.And finally,2kinds of abnormal operation of the landing phase are detected.