Dynamic Knowledge Inference and Learning for the Diagnosis of Insulator Running State Based on Fuzzy Petri Net

作者：Zhang, Zipeng^*; Wang, Shuqing

来源：Information-An International Interdisciplinary Journal, 2012, 15(9): 3803-3810.

• Fuzzy Petri net
• Expert system
• Neural network
• Diagnosis of insulator running state
• Adaptive learning

The working environment of insulator is complicated and there are many uncertain factors influence its running. For knowledge in expert reasoning system is limited and modified, it is very important to research the dynamic knowledge inference framework for expert reasoning system. An automatic fuzzy reasoning and learning framework based on fuzzy Petri net is presented for the diagnosis of insulator running state in this paper. The relative degree of each proposition of fuzzy Petri nets may be described accurately in the designed reasoning system. Continuous functions and hierarchy model are built to approximate transition firing and fuzzy reasoning in the design of reasoning system. In order to improve learning speed, the improved adaptive learning way is used to train the parameters of fuzzy reasoning rules in Petri net. Experiment result shows the designed reasoning system is effective in insulator running state diagnosis.

• 出版日期2012-9
• 单位湖北工业大学