Résultats de recherche pour « Low-Rank Dynamic Mode Decomposition »

Abstract This study presents an extended investigation of turbulent flow past a sphere at a Reynolds number of Re=1.6×104 using scale-resolving simulation (SRS) techniques. The first part of the work focuses on detailed numerical analysis of the flow physics employing the partially averaged Navier–Stokes (PANS) methodology derived from the k−ω Reynolds-averaged Navier–Stokes (RANS) formulation. Four resolution levels, characterized by the unresolved-to-total kinetic energy ratio (fk=0.1,0.2,0.3, and 0.5), were computed to examine wake development and near-wall turbulence characteristics. The resulting flow fields were analyzed through comparative visualization and spectral analysis to characterize the dominant coherent structures. The second part of the study introduces a reduced-order modeling (ROM) framework to extract and represent the principal dynamical features of the wake. Dynamic mode decomposition (DMD) is applied to identify energetically significant modes associated with the dominant instability mechanisms. A DMD is utilized to retain primary vortex shedding and Kelvin–Helmholtz motions. The combined high-fidelity simulations and data-driven decomposition provide insight into the multiscale dynamics of sphere wakes and demonstrate the capability of hybrid-scale resolving and reduced-order modeling strategies for complex turbulent flows.

Abstract Traffic forecasting is an important part in realising intelligent traffic management, which helps traffic controllers and travellers make effective decisions. However, traffic forecasting accuracy is often affected by missing traffic data due to hardware and software failure. Therefore, accurate prediction based on incomplete traffic data is an important problem as well as a challenge. Though many approaches recover the missing values before prediction, the errors from the data‐filling step are likely to cause additional bias to the prediction result. Besides, this tactic is difficult to guarantee the timeliness and may impede real‐time prediction. In this case, a traffic forecasting model is proposed to directly predict traffic data with missing values. This model develops tensor formed dynamic mode decomposition, recording the dynamic information of traffic data into a state transition tensor. In addition, the model takes low rank property of the dynamic tensor and the similarity of temporal variation trend into consideration. In order to verify the effectiveness and the robustness of the proposed model, experiments were performed on two real‐world time series datasets. The results demonstrate that the model achieves better performance on forecasting than other baseline approaches under the impact of missing data.