"Bond graph approach for input-output decoupling of linear MIMO systems with Derivative State Feedback"

Author(s):  
Christophe Sueur

"This paper presents a new solution for the well-known input-output decoupling problem of linear multivariable systems with a derivative state feedback control law. A simple solution to the pole placement problem is highlighted in the monovariable and multivariable cases with application to a mechanical system. Analysis up to control design are achieved structurally in the bond graph domain for the case study."

1999 ◽  
Vol 121 (1) ◽  
pp. 18-26 ◽  
Author(s):  
S. Y. Huang ◽  
K. Youcef-Toumi

Zero dynamics is an important feature in system analysis and controller design. Its behavior plays a major role in determining the performance limits of certain feedback systems. Since the intrinsic zero dynamics can not be influenced by feedback compensation, it is important to design physical systems so that they possess desired zero dynamics. In the Part I paper, a method is proposed to derive the zero dynamics of SISO systems from bond graph models. Using this approach, the design of physical systems, including the consideration of zero dynamics, can be performed in a systematic way. In this paper, the extension of the proposed method for MIMO systems is presented. It is shown that for MIMO systems, the input-output configurations determine the existence of vector relative degrees. If a system has a vector relative degree, it’s zero dynamics can be identified by a straightforward extension of the proposed method. If a system does not have a vector relative degree, a dynamic extension procedure may be used to fix the structure. By doing so, the zero dynamics can still be identified in a similar manner. It is also shown that if the input-output configurations are ill-designed, not only the relative degrees do not exist, but also the zero dynamics can not be reasonably defined. In that case, independent tracking controls for all the outputs are impossible. Therefore, the results in this paper provide a guideline for the design of the input-output configurations as well as the zero dynamics of MIMO systems.


1997 ◽  
Vol 119 (3) ◽  
pp. 478-485 ◽  
Author(s):  
M. Goldfarb ◽  
N. Celanovic

A lumped-parameter model of a piezoelectric stack actuator has been developed to describe actuator behavior for purposes of control system analysis and design, and in particular for control applications requiring accurate position tracking performance. In addition to describing the input-output dynamic behavior, the proposed model explains aspects of nonintuitive behavioral phenomena evinced by piezoelectric actuators, such as the input-output rate-independent hysteresis and the change in mechanical stiffness that results from altering electrical load. Bond graph terminology is incorporated to facilitate the energy-based formulation of the actuator model. The authors propose a new bond graph element, the generalized Maxwell resistive capacitor, as a lumped-parameter causal representation of rate-independent hysteresis. Model formulation is validated by comparing results of numerical simulations to experimental data.


Author(s):  
Pilla Ramana ◽  
Karlapudy Alice Mary ◽  
Munagala Surya Kalavathi

Control system design for inverter fed drives previously used the classical transfer function approach for single-input singleoutput (SISO) systems. Proportional plus Integral (PI) controllers were designed for individual control loops.It is found that the transient response of a PI controller is slow and is improved by pole placement through state feedback. However, the effective gains of the PI controller are substantially decreased as a function of the increase of motor speed. A control system is generally characterized by the hierarchy of the control loops, where the outer loop controls the inner loops. The inner loops are designed to execute progressively faster. The speed controller (PI controller) processes the speed error and generates the reference torque. In the inner loop, firstly a non-linear controller is designed for the system by which the system nonlinearity is canceled using state or exact feedback linearization. In addition, a linear state feedback control law based on pole placement technique including the integral of output error (IOE) is used in order to achieve zero steady state error with respect to reference current specification, while at the same time improving the dynamic response.The proposed scheme has been validated through extensive simulation using MATLAB.


2017 ◽  
Vol 29 (3) ◽  
pp. 591-601
Author(s):  
Ryota Hayashi ◽  
◽  
Genki Matsuyama ◽  
Hisanori Amano ◽  
Hitomu Saiki ◽  
...  

[abstFig src='/00290003/14.jpg' width='300' text='Amphibian vehicle maneuvering simulator' ] This study proposes a maneuvering support system for an amphibian vehicle by applying a nonlinear state feedback control law for vehicle trajectory control. We consider that the vehicle should not drift sideways for good driving performance. To derive a nonlinear state feedback control law, we have defined ‘Maneuvering Trajectory’ as an additional reference trajectory that is generated by the driver’s maneuver. We have constructed a Lyapunov-like function for the trajectory control system. In this paper, we construct a vehicle-maneuvering simulator and set a clockwise circular reference trajectory. The efficiency of the proposed maneuvering support system is shown in the maneuvering simulations. We consider the case where the propulsive forces of the vehicle have limited influence on maneuverability. A new warning display system is proposed so that the driver can recognize if his or her maneuver is not suitable. Then, we examine the feasibility of the proposed warning display system through several simulations.


1997 ◽  
Vol 122 (1) ◽  
pp. 49-62 ◽  
Author(s):  
A. S. Tsirikos ◽  
K. G. Arvanitis

The disturbance rejection with simultaneous input-output linearization and decoupling problem of nonsquare nonlinear systems via restricted state feedback is investigated in this paper. The problem is treated on the basis of an algebraic approach whose main feature is that it reduces the determination of the admissible state feedback control laws to the solution of an algebraic and a first order partial differential systems of equations. Verifiable necessary and sufficient conditions of algebraic nature based on these systems of equations are established for the solvability of the aforementioned problem. Moreover, an explicit expression for a special admissible restricted state feedback controller is analytically derived. [S0022-0434(00)02101-8]


Sign in / Sign up

Export Citation Format

Share Document