不确定条件下航空事故预警及预防对策生成机制研究

The uncertainty theory is introduced in the predicting, warning and prevention of the aviation accidents. The research would be done along the thoughts as information collecting, model building, accident predicting and preventing. This research is aiming to build a mathematic model of the aviatic safety and its uncertain influencing factors, proposed a more scientific and effective accdent's predicting, early warning method and generative mechanism of prevention measures, what’s more, the project could explore the fine management theory of the aviatic safety. .Firstly, considering the uncertainty, a more perfect index system of the aviatic safety are proposed according to the data mining work of the aviation safety, and the responding index is described by a series of uncertain theories. Secondly, the unsafe incidents, including accidents, accident symptoms, ground accidents etc., are analyzed, and their generative rules are obtained. Therefore, a more comprehensive index model, the synthesis safety index model, is put forward to comprehensively reflect the aviation safety, and a new predicting methodology which based on the information update theory is proposed to forecast the following aviation safety. Thirdly, using the reliability theory and probabilistic safety risk assessment theory, the quantitative mapping function between the aviation safety synthesis safety index model and uncertain indices is established and analyzed. Combining with the importance and reliability sensitivity analysis method, the importance and effects of various uncertain important factors are solved. Based on that, the formation mechanism of aviation security prevent measures are proposed. Finally, take an aviation unit as an example, the theory and methodology are applied to test their feasibility and validity.

将不确定性理论引入航空事故预测、预警、预防理论研究中，通过信息分析、模型构建、预测预警、对策生成的思路开展研究。项目力图提出一种科学有效的航空事故预测、预警方法及预防对策生成机制，为航空安全精细化管理作出有益探索。首先，对航空安全管理产生的信息实施数据挖掘，利用不确定理论对相应事故诱因进行描述，建立更加科学完善的航空安全影响指标体系；其次，利用所掌握不安全事件数据，研究航空事故的发生发展规律，提出能够全面反映航空安全的综合安全指数模型，并提出能够进行信息实时更新的航空安全预测、预警方法；再次，建立航空综合安全指数与不确定性因素之间的定量映射函数关系，研究各不确定因素对航空安全综合评价模型影响的重要程度及影响程度，研究不确定性条件下航空安全预防对策生成机制（包括思路、流程、验证方法等）。最后，以某航空单位为例，对所研究的理论方法进行实践应用。

项目将不确定性理论引入航空安全管理的预测、预警与预防研究中，重点探索航空安全中不确定信息的描述处理、预警指标模型的构建方法、预防对策的生成机制。首先，通过收集大量航空事故，梳理并提出分析影响航空安全的不确定因素，构建了航空不安全事件的诱因指标体系，利用不确定理论研究了航空安全中的不确定因素量化和描述方法。其次，根据航空不安全事件发生规律，构建了不同层次的航空安全综合评价指数模型，开展不确定条件下的航空安全综合评价模型和基于信息更新的航空安全预测方法研究。第三，在建立航空安全综合评价模型的基础上，研究了基于航空综合安全模型的重要度及灵敏度分析方法，探索了基于不确定性的航空安全预防对策生成机制，并对上述研究理论方法开展实践研究。特别地，项目针对航空装备类的复杂系统中的软件安全性问题数量激增的情况进行了深入研究，提出了一系列适用于航空装备的复杂系统软件安全性分析流程方法，并提出了用于军机机载软件的审查原则、方法，通过开发相关工具平台，获取了国家软件著作版权许可。