7990

7425

كارشناسي ارشد

كنترل

اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر

1391

ده،91ص.: جدول،نمودار

ص.ع.به فارسي و انگليسي

جواد عسگري، حميدرضا مرزبان

يداله ذاكري

9/7/92

محسن مجيري، مرضيه كمالي

مهندسي برق و كامپيوتر

ID7425

به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي

Explicit Model Predictive Control for Hybrid Systems with Application to Traction Control Roozbeh Memarzadeh r memarzadeh@ec iut ac ir r memarzadeh1982@yahoo com 2012 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Languege Persian Javad Askari j askari@cc iut ac ir Hamidreza Marzban hmarzban@cc iut ac ir Abstract MPC has become the accepted standard for complex constrained multivariable control problems in the process industries One of those contro ability to control a vehicle under adverse external conditions such as wet or icy roads By maximizing the vents the wheel from slipping and at the same time improves vehicle stability and steering ability In most control schemes the wheel slip i e the difference between the normalized vehicle speed and the speed of the wheel is chosen as the controlled variable The objective of the controller is to maximize the tractive torque while preserving the stability of the system The relation between the tractive force and the wheel slip is nonlinear and is a function of the road condition The presence of nonlinearities and constraints on one hand and the simplicity needed for real time implementation on the other have discouraged the use of optimal control theory for designing practical traction control systems In this thesis a multiparametric quadratic programming mp QP method is reviewed that moves all the computations necessary for the implementation of MPC off line while it preserves all other characteristics of the MPC controller The mp QP is successfully implemented both for reference tracking and state regulation problems Due to the nonlinear nature of the wheel slip dynamics a linear model is inadequate and standard MPC methodologies can not be applied to the traction control problem We show how a Mixed Logical Dynamical MLD model for the traction system can be obtained using the HYSDEL compiler Two efficient method are presented for computing the explicit MPC for MLD and Piecewise Affine PWA systems We apply our methodology to a piecewise affine model of the traction system and compute an explicit nonlinear traction control law with very good performance in simulations Key words Multiparametric Linear Programming Multiparametric Quadratic Programming Mixed Logical Dynamical system MLD Hybrid System Piecewise Affine system PWA HYbrid System DEscription Language HYSDEL PDF created with pdfFactory trial version www pdffactory com

جواد عسگري، حميدرضا مرزبان

يداله ذاكري

محسن مجيري، مرضيه كمالي