An intelligent framework for fault diagnosis in 89c51RD2 Microcontroller based system

• Main Authors: Kodavade, D.V., Apte, S.D.
• Format: Conference Proceeding
• Language: English
• Subjects: Artificial intelligence; Circuit faults; Decoding; Fault detection; Fault diagnosis; Fuzzy; Hardware; Inference; Inference mechanisms; Knowledge base; Microcontrollers; Testing; User interfaces
• DOI: 10.1109/IDAACS.2009.5342978

Fault diagnosis in digital hardware using AI techniques is good domain for basic and applied research. An intelligent frame work for fault detection and isolation in Philips 89v52 RD2 microcontroller based system is discussed in this paper. The main feature includes intelligent diagnostic assessment and effective management of the testing process using fuzzy approaches. Fuzzy modeling is used to derive nonlinear models for the diagnosis process for every fault. Fuzzy decision factors are derived to isolate faults. The unit under test (UUT) consist of 89c51-RD2 microcontroller, external memory and other peripherals like decoders, TTL gates etc. The framework consist of knowledge base, inference mechanism and graphical user interface. The knowledge base consist of three distinct parts such as, the experiential knowledge, fundamental knowledge and the symptoms. The knowledge base consist of a procedural description of the test, expressed in a hierarchical manner using visual Prolog rules and declarative knowledge represented using frames. The system uses both deep and shallow knowledge about troubleshooting process. The frame work performs inference in the similar manner in which an electronic engineer traces a logic circuit. The diagnosis determines the causes of the differences between a system's expected behavior and its observed behavior under same input vectors. The user interface is developed using visual programming aspects which graphically shows the diagnostic process carried out. The user interface permits the user to enter observed symptoms as and when required by the system.