The Realization Of Analog Circuit Simulation And Data Management Software

At present, analog circuit fault diagnosis based on pattern recognition is one of the major ways to diagnosis. In intelligent fault diagnosis based on pattern recognition technology, the diagnosis technology of artificial intelligence is various, such as BP, AIS, SVM, etc. Therefore, the fault diagnosis knowledge interface is different and diagnostic knowledge representation format relies heavily on diagnostic reasoning module. As the result of diagnostic reasoning module can only parse the diagnostic data in a specific format, so there is no uniform standard for the diagnostic knowledge and data. Then it becomes difficult to share diagnostic knowledge between fault diagnosis systems.To regulate diagnostic process and realize the sharing of diagnostic knowledge, the Artificial Intelligence Exchange and Service Tie to All Test Environments (AI-ESTATE) standard was developed by the Diagnostic and Maintenance Control Subcommittee of the IEEE Standards Coordinating Committee 20 (SCC20) on Test and Diagnosis for Electronic Systems. But in the AI-ESTATE standard there are only four kinds of diagnostic information model defined, which is used to standardize several common fault diagnosis methods. There is not the model specification to corresponding artificial intelligence such as BP, AIS, SVM algorithms which are often used in intelligent fault diagnosis based on pattern recognition technology. Through the research about AI-ESTATE model structure and BP neural network algorithm, this paper puts forward a new diagnostic information model--the BP neural network model (also known as BP information model). Based on BP information model, the prototype system of analog circuit simulation and data management software based on BP interface is developed, which implements diagnostic knowledge sharing and overcomes the disadvantages of diagnostic module poor portability and the diagnostic knowledge can’t be exchanged in traditional fault diagnosis system.First, the BP information model is designed by studying the common element model and the BP algorithm. Then this paper introduces the BP information model and describes the EXPRESS specification of each entity in model. And it analyzes the function and relationship of BP model each entity in the process of testing and fault diagnosis.Then, the BP model transformation module is designed by analysising the structure information and diagnosis process of the BP information model. It can implement the generation of BP exchange files. The BP exchange files contained the diagnostic process information and the fault diagnosis knowledge base, can be shared between diagnostic reasoners with standard interfaces. In order to complete the diagnostic process, the diagnostic reasoner is designed and implemented. The reasoner can guide the test and diagnosis process in the fault diagnostic system by analyzing the BP exchange files.Finally, the the prototype system of analog circuit simulation and data management software is realized. And the experiment is carried out by using the circuit board which is provided by the project. The validation results show that the BP information model based on AI-ESTATE standard is feasible.