Multibody Systems Containing Active Elements: Generation of Linearized System Equations, System Analysis and Order-Reduction
The nucleus of a general purpose computer program for the dynamics of vehicle-guideway-interaction is an algorithm which automatically generates the linearized equations of motion of arbitrary configured multibody systems connected by linear elements. Every body of the system can be rigid or flexible and arbitrary translational and rotational modes of motion can be allowed. Also, kinematic constraints are used for reducing the system order to the minimal number of (independent) degrees of freedom (DOF).
Besides of the usual basic coupling elements as linear spring-damper Systems (in parallel or in series), a broad class of more general elements can be modelled: dynamic element including active control devices (sensors, actuators and (dynamic) feedback).
The resulting (state-) equations can be evaluated directly with respect to a number of important qualitative and structural system properties such as stability, controllability, observability and transfer functions. It is recommended to base all computations on the eigenvalues and -vectors of the system matrix F.
for weak coupling and/or fast decaying eigenmodes to discard the associated transient dynamics, and
for strong coupling to use a condensation procedure based on the second-order multibody equations [approximate generalized kinematic constraints).
The main advantages are the correct reproduction of the steady-state values, the possibility of adaptation of the kept eigenvalues, as well as the close preservation of the physical structure of the models in connection with relative numerical simplicity.
The use of the prescribed procedures of model-synthesis, -analysis and -order-reduction is demonstrated on a typical levitated ground vehicle model (electrodynamic repulsion system), where the order can be reduced from more than 70 down to about 10 (!) states without considerable deterioration of accuracy.
KeywordsMultibody System Kinematic Constraint Order Reduction Linearize System Equation Magnetic Suspension
Unable to display preview. Download preview PDF.
- 2.Duffek, W.; Kortüm, W.; Wallrapp, O.: A General Purpose Program for the Simulation of Vehicle Guideway Interaction Dynamics. Paper to be presented at 2nd IUTAM Symposium on Dynamics of Vehicles, Vienna, Sept. 19–23, 1977-Google Scholar
- 3.Kortüm, W.; Lehner, M.; Richter, R.: Multibody Systems Containing Active Elements: Algorithmic Generation of Linearized System Equations, System Analysis and Order-Reduction. Internal DFVLR-report in preparation.Google Scholar
- 5.Davison, E.J.: A Method for Simplifying Linear Dynamic Systems. IEEE Transactions on Automatic Control, Vol. AC-11, No. 1, January 1966.Google Scholar
- 9.Kokotovic, P.V.; O’Malley, jr., R.E.; Sannuti, P.: Singular Perturbations and Order Reduction in Control Theory — An Overview. Automatica, Vol. 12, pp. 123–132, Pergamon Press, 1976.Google Scholar
- 10.Hutton, M.F.: Routh Approximation Method and Singular Perturbations. Proc. 13th Allerton Conf. on Circuit and System Theory, Univ. of Illinois, Urbana, Illinois, Oct. 1975-Google Scholar
- 11.Hutton, M.F.; Rabins M.J.: Simplification of High-Order Mechanical Systems Using the Routh Approximation. Journal of Dynamic Systems, Measurement, and Control, December 1975, pp. 383–392.Google Scholar