Research On Key Technologies Of Fault Diagnosis System Based On IETM

With the increasingly complex function,structure,and integration technology of self-propelled artillery,the requirements for fault diagnosis and maintenance support are also increasing.At the same time,the mobility of self-propelled artillery users is large,and it is essential to ensure the training and learning outcome of equipment and reduce the dependence of diagnostic reasoning on professional skills and experience,so that users and maintenance personnel can quickly familiarize themselves with equipment,and accurately and quickly apply knowledge to complete fault diagnosis and troubleshooting when faults occur.The application of IETM not only provides convenience for fault diagnosis of complex equipment,but also meets the needs of equipment maintenance and training.In order to improve the fault diagnosis efficiency and support ability of self-propelled artillery,this thesis studies the design and implementation method of self-propelled artillery fault diagnosis system based on IETM framework from three aspects: system model design,knowledge representation and storage,and diagnostic inference process.The main research contents include:(1)Aiming at the problem of how to implement a self-propelled artillery fault diagnosis system under the IETM framework,this thesis analyzes the design requirements,technical difficulties,and implementation methods of an IETM based fault diagnosis system based on the structure and fault information characteristics of self-propelled artillery.The system model is designed and the functional framework of the core modules of the system is given.(2)Aiming at the problem of insufficient atomicity subdivision of IETM data directly used for system diagnosis reasoning,a knowledge base of the system is designed based on IETM fault information data module.The knowledge acquisition and representation methods were provided,and a fault knowledge storage structure with fault codes,fault locations,and fault descriptions as the core was designed.And combine database technology to achieve the conversion of IETM data to the system knowledge base.The knowledge storage structure can associate data in IETM to provide data support for system diagnosis and inference.(3)The inference engine of the system is designed,and the diagnosis inference process is designed based on the knowledge storage structure.In order to exert the interactive ability of IETM and overcome the inherent limitations of pure machine reasoning,an interactive diagnostic reasoning process was designed.The method enhances the flexibility of reasoning by combining rule reasoning with human-computer interaction.At the same time,in order to solve the problem of difficult to determine the fault cause caused by the uncertain relationship between fault knowledge,a diagnostic inference process based on fuzzy fault tree and Bayesian network is designed.The uncertainty between faults is described by probability using fuzzy fault tree and Bayesian network technology.In the reasoning process,factors such as frequency,equipment structure level,and membership relationship between faults are introduced on the basis of the self-propelled artillery fault tree,and the fuzzy probability of the base event is calculated through fuzzy operators.Taking it as a prior probability,the Bayesian theorem is used to calculate the posterior probability of the root node in the Bayesian network.By comparing the posterior probability,the most likely cause of failure is determined.The method effectively reduces the impact of human subjective factors and meets the needs of inaccurate and rapid reasoning in the system.Finally,taking a self-propelled artillery as an example,a fault diagnosis prototype system is constructed,and the effectiveness of the model and method proposed in this thesis is verified through an example analysis.