Abnormality diagnosis system and abnormality diagnosis method

An abnormality diagnosis system for diagnosing an abnormality in industrial equipment, comprising:

• a data analysis unit that analyzes data based on status monitoring information of the industrial equipment;
• and
• a display device that displays an analysis result of the data,
• wherein the data analysis unit includes a validity determination unit that calculates a determination factor for determining validity of an analysis process of the data to determine the validity of the data analysis process,
• wherein the display device displays a determination result of the validity,
• wherein the data analysis unit further includes:
• a preprocessing unit that preprocesses data necessary for the analysis of the status monitoring information;
• and
• a feature amount extraction unit that extracts a feature amount as a feature of the analysis from the preprocessed data,
• wherein the validity determination unit calculates the determination factor for determining the validity based on input information from the preprocessing unit and the feature amount extraction unit,
• wherein the display device is configured by a display input device to which predetermined information can be input,
• wherein the data analysis unit further includes a diagnostic unit that diagnoses an abnormality in the industrial equipment by using the feature amount,
• wherein the diagnostic unit diagnoses an abnormality in the industrial equipment by using a diagnostic parameter and a diagnostic model input from the display input device,
• wherein, when the diagnostic model is adjusted by changing the diagnostic parameter, the diagnostic unit receives an adjustment range of the diagnostic parameter from the display input device in advance as a diagnostic parameter search width and diagnoses an abnormality in the industrial equipment,
• wherein the data analysis unit analyzes the data a plurality of times according to the diagnostic parameter search width, and
• wherein the display input device displays the validity determination result so as to correspond to the diagnostic parameter search width.

An abnormality diagnosis method for diagnosing an abnormality in industrial equipment, comprising:

• analyzing data based on status monitoring information of the industrial equipment;
• calculating a determination factor for determining validity of an analysis process of the data to determine the validity of the data analysis process;
• and
• displaying an analysis result of the data and a determination result of the validity on a display device that is configured by a display input device to which predetermined information can be input;
• preprocessing data necessary for the analysis of the status monitoring information;
• extracting a feature amount as a feature of the analysis from the preprocessed data;
• calculating the determination factor for determining the validity based on input information from the preprocessing of the data and the extracting of the feature amount;
• diagnosing an abnormality in the industrial equipment by using the feature amount;
• diagnosing an abnormality in the industrial equipment by using a diagnostic parameter and a diagnostic model input from the display input device;
• and
• storing, in a determination database, a method for correcting the diagnostic parameter corresponding to the validity determination result,
• wherein the method for correcting the diagnostic parameter corresponding to the validity determination result is read out from the determination database and displayed on the display device,
• wherein the display device displays, as the validity determination result, whether or not the data analysis is valid or not for each determination item,
• wherein when the data analysis is not valid, the diagnostic parameter correction method corresponding to the validity determination result is read out from the determination database, and
• wherein the diagnostic parameter correction method is displayed so as to correspond to the determination item for which the data analysis is not valid.