.png){kind=icon}
.png){kind=icon}
Return
On Train Dynamics Modeling Based on Parameter Fuzzy Identification Method
.png){kind=icon}
PRE
AI
DOI：10.1007/s40815-025-02001-4.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Aiming at the dynamic modeling problem of high-speed train under complex operating environment, this paper considers the uncertainty of nonlinear parameters in the drag function, the saturation constraint of traction or braking force, and the key factors such as the speed delay caused by sudden factors. Variable Forgetting Factor Recursive Least Squares based on Fuzzy Rules (VF-RLS) is proposed. The capture and identification of nonlinear parameters in the resistance function of dynamic model are realized. Based on the fuzzy rules designed by expert experience, the forgetting factor is adjusted in real time to estimate the time-varying three coefficients of the nonlinear total resistance function and the train mass. In this paper, 5228 sets of tractor-speed train running data of a certain line are used as input data to verify the algorithm simulation. The results show that VF-RLS improves the accuracy by 16% to 45% compared with traditional parameter identification methods.
Keywords:
Single particle dynamics model
Parameter identification
Recursive least squares method
Variable forgetting factor
Fuzzy rules
Journal
IF:
3.6
Papers:
2.2K
Citations:
4.3K