Method for detecting abnormal vibration of rolling mill, apparatus for detecting abnormality of rolling mill, rolling method, and method for producing metal strip
A method for detecting abnormal vibration of a rolling mill including a collecting step of collecting vibration data of the rolling mill, a frequency analysis step of generating first analysis data by performing frequency analysis of the vibration data, a principal component analysis step of performing principal component analysis on the first analysis data by using reference data specified in advance on the basis of a normal state as a principal component and thereby generating evaluation data, which is a projection of the first analysis data onto the reference data, and an abnormal vibration detection step of extracting an outlier component from the evaluation data and the first analysis data and detecting an abnormality of the rolling mill from the extracted outlier component.
1 . A method for detecting abnormal vibration of a rolling mill that has a pair of work rolls and a plurality of backup rolls that support the work rolls, the method comprising:
a collecting step of collecting vibration data of the rolling mill;
a frequency analysis step of generating first analysis data by performing frequency analysis of the vibration data;
a principal component analysis step of performing principal component analysis on the first analysis data by using reference data specified in advance on a basis of a normal state as a principal component and thereby generating evaluation data, which is a projection of the first analysis data onto the reference data; and
an abnormal vibration detection step of extracting an outlier component from the evaluation data and the first analysis data and detecting an abnormality of the rolling mill from the extracted outlier component, wherein
in the principal component analysis step, the principal component extracted as the reference data is set for each rolling speed in the rolling mill.
2 . The method according to claim 1 , wherein the frequency analysis step generates vibration intensities at respective frequencies as the first analysis data,
the method further comprises a data conversion step of converting the first analysis data into second analysis data indicative of vibration intensities at respective intervals on a basis of a rolling speed, and
the principal component analysis step performs principal component analysis of the second analysis data.
3 . The method according to claim 1 , wherein a plurality of principal components extracted as the reference data in the principal component analysis step are set so that a cumulative value of contribution ratios of the principal components becomes equal to or larger than a reference contribution ratio when principal component analysis is performed on normal analysis data acquired during rolling performed by the rolling mill in a normal state.
4 . The method according to claim 1 , wherein the rolling mill cold-rolls a metal strip.
5 . A rolling method comprising a backup roll replacement step of, in a case where an abnormality of a rolling mill is detected by using the method according to claim 1 , replacing the backup rolls of the rolling mill.
6 . A method for producing a metal strip, the method comprising a step of producing a metal strip by using the rolling method according to claim 5 .
7 . The method according to claim 2 , wherein a plurality of principal components extracted as the reference data in the principal component analysis step are set so that a cumulative value of contribution ratios of the principal components becomes equal to or larger than a reference contribution ratio when principal component analysis is performed on normal analysis data acquired during rolling performed by the rolling mill in a normal state.
8 . A rolling method comprising a backup roll replacement step of, in a case where an abnormality of a rolling mill is detected by using the method according to claim 2 , replacing the backup rolls of the rolling mill.
9 . A method for producing a metal strip, the method comprising a step of producing a metal strip by using the rolling method according to claim 8 .
10 . An apparatus for detecting an abnormality of a rolling mill that has a pair of work rolls and a plurality of backup rolls that support the work rolls, the apparatus comprising:
a data collecting unit that collects vibration data of the rolling mill;
a frequency analysis unit that generates first analysis data by performing frequency analysis of the vibration data;
a principal component analysis unit that performs principal component analysis on the first analysis data by using reference data specified in advance on a basis of a normal state as a principal component and thereby generates evaluation data, which is a projection of the first analysis data onto the reference data; and
an abnormality detection unit that extracts an outlier component from the evaluation data and the first analysis data and detects an abnormality of the rolling mill from the extracted outlier component, wherein
the principal component extracted as the reference data by the principal component analysis unit is set for each rolling speed in the rolling mill.