Detecting Stochastic and Deterministic Structures of Short-Term Metro Passenger Flow with CEEMDAN and RQA

Considering the nonlinear, non-stationary, and chaotic characteristics of the short-term passenger flow, this paper proposes a hybrid model combing the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and recurrence quantification analysis (RQA), to detect the stochastic and deterministic structures of short-term passenger flow. In the model, CEEMDAN is used to decompose the original data into several intrinsic mode functions and a residue, while RQA is performed to reconstruct the decomposed modes into a stochastic part, a deterministic part and a trend part via determinism evaluation. Further, RQA and a windowed RQA are conducted to analyze the static and time-evolving dynamic behaviors of the reconstructed components. The data from Chengdu metro system, China, is employed to verify the proposed model. The results suggest that the proposed model provides an effective way to detect the intrinsic dynamics and uncover underlying structures of the short-term metro passenger flow.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; References; Tables;
  • Pagination: 18p

Subject/Index Terms

Filing Info

  • Accession Number: 01764428
  • Record Type: Publication
  • Report/Paper Numbers: TRBAM-21-03957
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 23 2020 11:05AM