Research On Network-Based Remote Fault Diagnosis Expert System For NC Device

Monitoring and diagnostic technology for industrial process and equipments is an important means and key technology in ensuring running secure and reliability of manufacture system and in improving products' quality and production efficiency of the modern industry.The manufacturing industry has generally moved towards the integration, the network,the globalization development direction.It is important that the complex numerical control devices in the industrial production sector can operate in good state so as to improve the enterprise economic benefits,even the comprehensive national strength. Traditional fault diagnosis technology can not adapt to the requirements of fault diagnosis for complex devices.Therefore,it is necessary to research and develop new effective and accurate method of fault diagnosis.According to the isolated production pattern of numerical control device in the machine shop,the multiple and maintenance difficult of fault and the development trend of fault dignosis technology,the present dissertation developed a network-based remote fault diagnosis system to prevent fault occurrence and quicken the recovering process.We take NC device as the fault diagnosis research object and detailed analyze the functional requirements of remote monitoring and fault diagnosis system.The article designs the overall framework of remote fault diagnosis system and the hardware circuit of communication module and addresses relevant application technology of the fault diagnosis system.In order to solve the defects of rules matching conflict,slow search speed and low reasoning efficiency of traditional fault dignosis expert system which is based on rules,the article diagnoses the faults of numerical control device with the combination of the expert system and the Petri nets theory,and emphasizes on the establishment of fuzzy expert system based on the object-oriented and visualization thought.An improved fuzzy Petri nets concurrent reasoning mechanism is proposed to reduce the iteration times and computational complexity in each iteration process of original algorithm.The improved fuzzy Petri nets concurrent reasoning algorithm,which possesses the capability of uncertain knowledge and parallel processing,is applied to handle the fuzziness and concurrency of fault and inadequate and uncertain information. Utilization of simple matrix operation is applied to realize complicated reasoning process that makes the diagnostic reasoning decision-making process visualized and simplify.A practical fault instance of numerical control device is analyzed to demonstrate the feasibility and validity of the proposed fuzzy Petri nets concurrent reasoning algorithm, which is implemented by using Matlab language.Finally,fault diagnosis system,which has realized the fault query function,is developed by VC++ and SQL 2000 softwares.