Group-sma algorithm based joint estimation of train parameter and state

Wei Zheng; Juan Han; Weijie Kong; Lixiang Wang

doi:10.7307/ptt.v27i1.1499

Group-sma algorithm based joint estimation of train parameter and state

Wei Zheng^*
, Juan Han
, Weijie Kong
, Lixiang Wang

^*Corresponding author for this work

Beijing Jiaotong University

Research output: Contribution to journal › Article › peer-review

Abstract

The braking rate and train arresting operation is important in the train braking performance. It is difficult to obtain the states of the train on time because of the measurement noise and a long calculation time. A type of Group Stochastic M-algorithm (GSMA) based on Rao-Blackwellization Particle Filter (RBPF) algorithm and Stochastic M-algorithm (SMA) is proposed in this paper. Compared with RBPF, GSMA based estimation precisions for the train braking rate and the control accelerations were improved by 78% and 62%, respectively. The calculation time of the GSMA was decreased by 70% compared with SMA.

Original language	English
Pages (from-to)	85-95
Number of pages	11
Journal	Promet - Traffic and Transportation
Volume	27
Issue number	1
DOIs	https://doi.org/10.7307/ptt.v27i1.1499
State	Published - 2015
Externally published	Yes

Keywords

Parameter estimation
Particle filter
Rail braking system
State estimation

Access to Document

10.7307/ptt.v27i1.1499

Cite this

@article{48faecb92b7a46dd8d08336ea0f33082,

title = "Group-sma algorithm based joint estimation of train parameter and state",

abstract = "The braking rate and train arresting operation is important in the train braking performance. It is difficult to obtain the states of the train on time because of the measurement noise and a long calculation time. A type of Group Stochastic M-algorithm (GSMA) based on Rao-Blackwellization Particle Filter (RBPF) algorithm and Stochastic M-algorithm (SMA) is proposed in this paper. Compared with RBPF, GSMA based estimation precisions for the train braking rate and the control accelerations were improved by 78\% and 62\%, respectively. The calculation time of the GSMA was decreased by 70\% compared with SMA.",

keywords = "Parameter estimation, Particle filter, Rail braking system, State estimation",

author = "Wei Zheng and Juan Han and Weijie Kong and Lixiang Wang",

year = "2015",

doi = "10.7307/ptt.v27i1.1499",

language = "英语",

volume = "27",

pages = "85--95",

journal = "Promet - Traffic and Transportation",

issn = "0353-5320",

publisher = "Faculty of Transport and Traffic Engineering",

number = "1",

}

TY - JOUR

T1 - Group-sma algorithm based joint estimation of train parameter and state

AU - Zheng, Wei

AU - Han, Juan

AU - Kong, Weijie

AU - Wang, Lixiang

PY - 2015

Y1 - 2015

N2 - The braking rate and train arresting operation is important in the train braking performance. It is difficult to obtain the states of the train on time because of the measurement noise and a long calculation time. A type of Group Stochastic M-algorithm (GSMA) based on Rao-Blackwellization Particle Filter (RBPF) algorithm and Stochastic M-algorithm (SMA) is proposed in this paper. Compared with RBPF, GSMA based estimation precisions for the train braking rate and the control accelerations were improved by 78% and 62%, respectively. The calculation time of the GSMA was decreased by 70% compared with SMA.

AB - The braking rate and train arresting operation is important in the train braking performance. It is difficult to obtain the states of the train on time because of the measurement noise and a long calculation time. A type of Group Stochastic M-algorithm (GSMA) based on Rao-Blackwellization Particle Filter (RBPF) algorithm and Stochastic M-algorithm (SMA) is proposed in this paper. Compared with RBPF, GSMA based estimation precisions for the train braking rate and the control accelerations were improved by 78% and 62%, respectively. The calculation time of the GSMA was decreased by 70% compared with SMA.

KW - Parameter estimation

KW - Particle filter

KW - Rail braking system

KW - State estimation

UR - https://www.scopus.com/pages/publications/84937573263

U2 - 10.7307/ptt.v27i1.1499

DO - 10.7307/ptt.v27i1.1499

M3 - 文章

AN - SCOPUS:84937573263

SN - 0353-5320

VL - 27

SP - 85

EP - 95

JO - Promet - Traffic and Transportation

JF - Promet - Traffic and Transportation

IS - 1

ER -

Group-sma algorithm based joint estimation of train parameter and state

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this