| 摘要 |
PROBLEM TO BE SOLVED: To provide a diagnosis system of electric installations capable of diagnosing electric installations, by detecting noises and vibrations generated from the electric installations and applying the hidden Markov model to the detected noises and vibrations. SOLUTION: The diagnosis system 1 of electric installation is equipped with a sensor 3 for detecting noise and vibrations generated from the electric installation 2, a learning section 5 for memorizing preliminarily each parameter as diagnostic information, by parameterizing each pattern, where data obtained from the detected noises and vibrations are patterned, based on the hidden Markov model, a diagnostic section 6 in which generative probability of the patterns, where data obtained based on noises and vibrations detected from the sensor 3 during the operating state of the electric installation are patterned, is carried out an operation based on each parameter memorized preliminarily in the leaning section 5 and then electric installation 2 is diagnosed based on diagnostic information corresponding to the parameter, where the generative probability concerned becomes the maximum, and a display section 7 displaying the diagnostic results. COPYRIGHT: (C)2006,JPO&NCIPI
|
