Realtime selflearning method of diagnostic knowledge in intelligent diagnosis system
Article
Figures
Metrics
Preview PDF
Reference
Related
Cited by
Materials
Abstract:
Aiming at realtime diagnostic knowledge accumulation and updating in intelligent diagnostic system, to simulate the process experts accumulate and update fault diagnostic knowledge, a new realtime diagnostic knowledge selflearning model is proposed based on pattern comparing and updating. Abnormal distribution test of hypothesis is used to compare realtime equipment fault pattern to equipment normal pattern of contingent knowledge. The algorithm of sample size estimation algorithm is used to calculate the number of samples which is used to obtain necessary diagnostic efficiency. Realtime model updating algorithm is used to adjust the diagnostic knowledge model to actual equipment on the spot. Analysis to actual test data of equipment shows that the method could achieve new diagnostic knowledge accumulation and be adaptive to actual equipment on the spot.
