摘要
空中交通管制安全是确保飞行安全的重要因素。为提高空中交通管制系统的安全性,依据风险管理理论,建立了空中交通管制风险预警模型。根据SHEL模型对风险因素进行分析,建立了风险预警指标体系,在指标体系基础上引入神经网络评估方法,构建了空中交通管制风险预警模型,最后选取某基层空管机构数据样本,实现了预警模型的训练与检测。实例表明,预警结果与实际评价结果相吻合,达到了所需的精度。
Based on the study of the characteristic features of air traffic control system, this paper is aimed to come up with a renovated method of air traffic safety control risk evaluation based on the SHEL model. As a matter of fact, with the fast development of civil aviation industry, it has become a more and more urgent demand to provide effective and efficient methods for early warning methods for the safety of the flight. What we would like to propose in this paper is a new method known as the SHEL model and BP neural network for the air traffic safety control. In doing so, in this paper, we have first of all analyzed the influential factors in the air traffic control system, which can be roughly divided into 4 categories, that is, the human-hardware factor, human-software factor, human-environment factor and that of human-human involved factor. In our analysis, we would like to explore the relation between the risk factors and warning indexes of the air traffic control system, which can be chosen as the input parameters of BP neural network. And then, on the basis of the exploration of the early warning system, we have worked out a warning model of the air traffic safety control system via the BP neural network, which should be able to meet the demands for the function approximation, pattern recognition and evaluation decisions. Such a network system has to be trained and evaluated by using this model so that an integrated evaluation procedure of the safety air traffic control system can be assured. Last of all, we have chosen an air traffic management office as our case study embodiment. The risk evaluation results of the method shows that what we have proposed proves to be feasible and effective for the safety need of the air traffic control system.
出处
《安全与环境学报》
CAS
CSCD
北大核心
2014年第3期138-141,共4页
Journal of Safety and Environment
基金
国家自然科学基金项目(71171199)
