An Improved Robust Principal Component Analysis Model for Anomalies Detection of Subway Passenger Flow

Figure 8

In the two figures, the horizontal axis is time interval and the vertical axis is anomaly flow. (a) shows the exit anomaly flow of multiple stations on August 7; lines represent different stations. From 17:00 to 20:00, there are continuous negative and positive anomalies, because there was a sudden heavy rain at 17:00 that resulted in the passenger flow peak moving back. (b) shows the anomaly flow of Olympic Sports Center Subway Station near the Bird’s Nest stadium. It is obvious that the exit flow surged before the beginning of game and the entry flow surged after the end of game.

(a)

(b)