Research And Realization Of Expert Systems For Failure Diagnoses Based On JESS

With the development of modern mass production and scientific and technologicalprogress, the industry structure is the increasing complexity of the functions become moreperfect, more and more high degree of automation. The fault diagnosis technique is to ensurethat the system is efficient, an important means of safe, reliable operation, more attention hasbeen paid, the corresponding research is more and more. To monitor the running state of thesystem failure at the same time when the system is running to timely treatment and diagnosiscan save huge maintenance costs. At the same time, the high demands of the traditional faultdiagnosis technology has been the fault diagnosis of complex systems out of line, web-basedprogram into the inevitable trend of research to the field of fault diagnosis. The same time,from the point of view of software engineering, research and development of fault diagnosissystem also showed the trend of complicated software stability, scalability and compatibilityrequirements have also increased, the traditional software development methods encounteredincreasingly strong development bottleneck. By the method of fault diagnosis, the content anddevelopment trends of popular Java development framework for J2EE production ruleframework JESS open source neural network framework Neuroph.The main content is as follows:First, the study fault diagnosis on the basis of the development process and the theoryand application of various methods in the field of fault diagnosis, remote fault diagnosisexpert system as the main objective of the study;Second, analysis of the characteristics of remote fault expert system built using the J2EEframework, a set of expert system to support remote access;Thirdly, the existing rules engine in-depth study comparison, the final decision JESS ruleengine written in Java as a core part of the expert system, perfect integration with J2EEframework;Fourth, for the parameter prediction in fault diagnosis and monitoring were the work ofthe research and analysis of the effect of contrast of the neural network and ARMA time seriesmodel in the diesel engine parameter warning, in-depth study of the framework of Neurophopen source neural network, usingas the neural network diagnostic part of the core.