Studies On Theory And Strategy Of Intelligent Fault Diagnosis For Liquid-Propellant Rocket Engine

In this thesis the theory and method of intelligent fault diagnosis for a large turbo-pump fed liquid-propellant rocket engine (LRE) is innovatively proposed and developed,based on the hybrid reasoning strategy and the hybrid knowledge models including mathematical model,logic model,graphic model and qualitative model. The theory and method are constructed in the way of intelligent modeling technique and creative application of knowledge engineering,and focus on the difficult problems and critical techniques such as the unified treatment of LRE knowledge (experience,facts,rules,graph,system structure,behavior,model knowledge and measured data),the integration and translation of qualitative and quantitative knowledge,and the knowledge-based intelligent fault diagnosis reasoning methods. Firstly,besides the correlation of different knowledge,the concept and the types of LRE diagnosis knowledge are systematically described. Secondly,the graphic and modeling ways of different knowledge acquisition,such as qualitative and quantitative knowledge,shallow and deep knowledge,fuzzy and dynamic knowledge,are discussed and studied. Thirdly,the object-oriented way of LRE knowledge representation for causality,system structure,and models is presented,and an organization model of LRE diagnosis knowledge for integration of deep and shallow knowledge based on the LRE component structure and in the form of node knowledge base is given. And finally,the LRE intelligent fault diagnosis theory and strategy,which utilize synthetically different kinds of diagnosis knowledge and multiple reasoning methods,such as rule-based,model-based,fuzzy-knowledge-based,and dynamic-knowledge-based reasoning methods,are discussed and developed. The results verified by the test-firing data and the simulated fault samples show that,the theory and methods proposed give a theoretical basis for the design and realization of practical LRE fault diagnosis and health monitoring systems,and can be available for reference in the development of health monitoring technique for reusable space transportation systems in the future.