Extended-State-Observer-Based Double-Loop Integral Sliding-Mode Control of Electronic Throttle Valve

被引：98

作者：

Li, Yongfu ^{[1
,2
,3
]}

Yang, Bin ^{[1
,2
]}

Zheng, Taixiong ^{[1
,2
]}

Li, Yinguo ^{[1
,2
]}

Cui, Mingyue ^{[4
]}

Peeta, Srinivas ^{[5
,6
]}

机构：

[1] Chongqing Univ Posts & Telecommun CQUPT, Coll Automat, Chongqing 400065, Peoples R China

[2] Chongqing Univ Posts & Telecommun CQUPT, Ctr Automot Elect & Embedded Syst, Chongqing 400065, Peoples R China

[3] Chongqing Univ Posts & Telecommun CQUPT, Elect Informat & Networking Res Inst, Chongqing 400065, Peoples R China

[4] Nanyang Normal Univ, Coll Phys & Elect Engn, Nanyang 473061, Peoples R China

[5] Purdue Univ, Sch Civil Engn, W Lafayette, IN 47907 USA

[6] Purdue Univ, NEXTRANS Ctr, W Lafayette, IN 47907 USA

来源：

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS | 2015年 / 16卷 / 05期

基金：

中国国家自然科学基金;

关键词：

Double-loop integral sliding-mode control (DLISMC); extended state observer (ESO); electronic throttle (ET) control; INTERNAL-COMBUSTION ENGINES; CONTROL STRATEGY; KALMAN FILTER; SENSORLESS CONTROL; SYNCHRONOUS MOTOR; NEURAL-NETWORKS; SYSTEMS; DRIVES; COMMUNICATION;

D O I：

10.1109/TITS.2015.2410282

中图分类号：

TU [建筑科学];

学科分类号：

0813 ;

摘要：

An extended-state-observer-based double-loop integral sliding-mode controller for electronic throttle (ET) is proposed by factoring the gear backlash torque and external disturbance to circumvent the parametric uncertainties and non-linearities. The extended state observer is designed based on a non-linearmodel of ET to estimate the change of throttle opening angle and total disturbance. A double-loop integral sliding-mode controller consisting of an inner loop and an outer loop is presented based on the opening angle and opening angle change errors of ET through Lyapunov stability theory. Numerical experiments are conducted using simulation. The results show that the accuracy and the response time of the proposed controller are better than those of the back-stepping and sliding mode control.

引用

页码：2501 / 2510

页数：10

共 45 条

[1] Modified Recursive Least Squares algorithm to train the Hybrid Multilayered Perceptron (HMLP) network [J].