21-04-2023 дата публикации
Номер: CN115994315A
Принадлежит:
The invention relates to the technical field of railway track irregularity disease identification, in particular to a railway track irregularity potential disease identification method based on feature clustering, and the method comprises the following steps: S1, integrating track irregularity historical dynamic detection data of a target section; s2, establishing a track irregularity potential disease identification index system; s3, extracting a disease identification index value from the historical dynamic detection data; s4, standardizing the disease identification index value to obtain a to-be-clustered sample data set; s5, based on a k-means + + algorithm, calculating a KL clustering index of the sample data set; s6, judging whether diseases exist in the target section or not, if not, prompting that no diseases exist, and if yes, executing S7; and S7, classifying the sample data set into two classes, extracting position information of abnormal samples, and marking and completing potential ...
Подробнее