X-ray tube predictive fault indicator sensing device, X-ray tube predictive fault indicator sensing method, and X-ray imaging device

An X-ray tube predictive fault indicator sensing method, comprising:

• a mode setting process of setting any one of operation modes including a prohibition mode, a learning mode, and a predictive fault indicator sensing mode;
• a vibration data acquisition process of acquiring vibration data regarding vibration generated from an X-ray tube, and outputting a vibration data acquisition completion notification whenever the number of acquired vibration data items reaches multiples of a predetermined number of data items used for frequency analysis performed once;
• a frequency analysis process of performing the frequency analysis on each of vibration data items of the predetermined number acquired through the vibration data acquisition process;
• a state data acquisition process of acquiring state data indicating an operation state of the X-ray tube, and synchronizing the acquired state date with a timing at which the vibration data acquisition completion notification output through the vibration data acquisition process is received;
• a learning process of performing cluster analysis by using, as input data, a plurality of learning data items formed of frequency component data obtained through frequency analysis in the frequency analysis process and the state data synchronized through the state data acquisition process in a case where the learning mode is set through the mode setting process, and generating one or more cluster data items;
• an abnormality calculation process of calculating, as abnormality, a minimum distance among distances to surfaces of respective clusters generated through the learning process from a position indicated by predictive fault indicator sensing target data in a case where the predictive fault indicator sensing mode is set through the mode setting process, the predictive fault indicator sensing target data being formed of the frequency component data obtained through the frequency analysis in the frequency analysis process and the state data synchronized through the state data acquisition process;
• and
• a predictive fault indicator determination process of comparing the abnormality calculated through the abnormality calculation process with a predetermined threshold value so as to determine a predictive fault indicator.