A civil aviation emergency causal relationship analysis method based on a Bayesian network

Abstract

The invention discloses a civil aviation emergency causal relationship analysis method based on a Bayesian network. The method comprises the steps of firstly establishing a civil aviation emergency management field ontology as a Bayesian network model, and dedarizing the Bayesian network to obtain a Markov network; and then initializing the Markov network and iteratively optimizing the Markov network, taking the Markov network optimized in the last iteration as an optimal Markov network, and using the optimal confidence coefficient of each node in the optimal Markov network to evaluate the probability of the causal relationship of the civil aviation emergency. According to the method, the V-shaped structure in the Bayesian network is changed, the message transmission rate in the Markov network is increased in the reasoning process, and the causal relationship calculation complexity in a large-scale network is reduced; in the Markov network reasoning process, a cyclic confidence coefficient propagation algorithm is improved in combination with a differential privacy model, and the leakage of sensitive information during causal relationship reasoning in the civil aviation field is avoided.

Description

technical field [0001] The invention belongs to the technical field of emergency management of civil aviation emergencies, in particular to a method for analyzing the causality of civil aviation emergencies based on Bayesian networks. Background technique [0002] With the continuous development of the country and the civil aviation industry, the number of civil aviation emergencies has increased rapidly, and the types of incidents are diversified. For example, the accident of China Eastern Airlines Flight 586 (9.10MU586) in 1998 was the first civil aviation emergency landing incident in China; The September 11 terrorist attack in the United States (911) that occurred in 2001 was defined as a civil aviation terrorist attack for the first time; the Air China "4.15" Korean Busan air crash that occurred in 2002 (4.15CA129); the 8.24 Yichun air crash in Heilongjiang, China that occurred in 2010 ( 8.24VD8387); the Malaysia Airlines 370 incident (MH370) in 2014 and the 10.11 Hongq...

AI Technical Summary

Problems solved by technology

At present, due to the wide variety of civil aviation emergencies, the ontology in the field of civil aviation emergency management is constructed as a Bayesian network model for reasoning and research on the probability distribution of the causal relationship of civil aviation emergencies. The Bayesian network scale is large and the reasoning process High computational complexity, low reasoning efficiency, and disclosure of sensitive civil aviation information

Images