Research On The Knowledge Acquisition And Represention Of Aviation Engine Airworthiness

Airworthiness of the engine is run through its research and development-use-maintenance of thewhole process,to ensure the safety of the engine, reducing engine failure-prone production, in thedevelopment process must be strictly in accordance with airworthiness standards for the developmentand validation.Therefore,use intelligent means acquire domain knowledge,experience knowledge,expertise knowledge,which exist in the airworthiness certification specifications,guidelines,experience,tools, manuals,guides,and test data in the development of aircraft engine and build aircraftengines airworthiness knowledge base.For the development of China’s large aircraft engine,the engineairworthiness certification, training of pilots and other aviation security work is of great significance.This paper presents a hybrid knowledge of the fact that knowledge, an instance of knowledge,knowledge of causality, hierarchy knowledge of the introduction of object-oriented, and given to theimplementation process and examples. The knowledge is accurate, comprehensive expressed, and tomeet the ease of understanding, planning, expansion and access requirements; this paper analysis theoverall framework of knowledge-based systems，to achieve knowledge browse, add, delete, modify,query, and the consistency and integrity checks. due to the scalability of the system, its integrity andnormative in the follow-up system improvement reflected.Based SMOTE re-sampling method，wedesigned to definition extraction based on over-sampling using distance distribution information ofinstances.The experimental analysis shows that the re-sampling algorithm combines balanced randomforest constructed by the C4.5decision tree balance, the definition of dealing with knowledge ofaero-engine airworthiness extract this classification problem is highly unbalanced data sets, it canimprove the classification results, better than direct classification method in the whole corpus.