针对飞行器任务评估中数据种类多，数据量大，实时性强，评估过程复杂等问题，提出一种混合推理模式的飞行器评估系统。该系统包括多个功能独立的模块，能够完成从飞行器原始数据到最终评估结论的推理和寻因。为了满足实时性需求，提出评估树结构和触发点机制。根据飞行器评估的特点，在顶层事实获取阶段采用范例推理，充分利用其非精确推理和自学习的优点；在评估推理阶段，采用高效精确的产生式推理机制。历史数据的仿真测试表明该系统能够即时完成飞行器评估，评估过程全程自动化。%To solve the problem in the aerocraft assessment task, which includes the amount of data is huge, the assessment process is complex, and the conclusion of the assessment is inaccurate, an aircraft assessment system based on hybrid reasoning model is proposed. The system comprises a plurality of independent module on function, and completes reasoning and searching results from the original data of aerocraft to final assessment conclusion. This paper presents assessment tree and trigger point mechanism, in order to meet real-time requirements. According to the characteristics of the aircraft assessment, the top-facts are get using case-based reasoning to fully utilize its non-precise reasoning and self-learning advantages; reasoning in the assessment stage is use of accurate and efficient rule-based inference mechanism. Practical examples using historical data show that, the proposed assessment system performs very well, and can real-time complete aerocraft assement, and the whole process is automatic.
展开▼
