Riccati Transfer Matrix Method for Linear Tree Multibody Systems

2016 ◽  
Vol 84 (1) ◽  
Author(s):  
Junjie Gu ◽  
Xiaoting Rui ◽  
Jianshu Zhang ◽  
Gangli Chen ◽  
Qinbo Zhou

The Riccati transfer matrix method (RTMM) improves the numerical stability of analyzing chain multibody systems with the transfer matrix method for multibody systems (MSTMM). However, for linear tree multibody systems, the recursive relations of the Riccati transfer matrices, especially those for elements with multiple input ends, have not been established yet. Thus, an RTMM formulism for general linear tree multibody systems is formulated based on the transformation of transfer equations and geometrical equations of such elements. The steady-state response under harmonic excitation of a linear tree multibody system is taken as an example and obtained by the proposed method. Comparison with the finite-element method (FEM) validates the proposed method and a numerical example demonstrates that the proposed method has a better numerical stability than the normal MSTMM.

Author(s):  
Qicheng Zha ◽  
Xiaoting Rui ◽  
Feifei Liu ◽  
Hailong Yu ◽  
Jianshu Zhang

Transfer Matrix Method for Multibody Systems (MSTMM) has the advantages of no need to establish the global system dynamics equations, low order of the system matrix, high programming, and fast calculation speed compared to the ordinary dynamics methods. In this paper, the topological graph of the dynamics model, transfer equations, transfer matrix of overall system and the simulation program of dynamics of the self-propelled artillery system are established by using the new version of the transfer matrix method for multibody systems and the automatic deduction theorem of overall transfer equation of systems. Realize the rapid calculation of the deviation of the pitch angle and the revolution angles of the turret versus time in the self-propelled artillery. It provides a theoretical basis and simulation means for the dynamics analysis of the self-propelled artillery.


2019 ◽  
Vol 86 (6) ◽  
Author(s):  
Jianshu Zhang ◽  
Xiaoting Rui ◽  
Junjie Gu

The transfer matrix method for linear multibody systems is capable of providing precise solutions for the dynamics of various mechanical systems, but it may also suffer from numerical instability in some cases, where serial chains with a large number of mechanical elements are involved or high-frequency harmonic responses are computed. Combining such a transfer strategy with the Riccati transformation yields the Riccati transfer matrix method (RTMM), which can help improve the numerical stability. According to the existing method, the conventional transfer matrices of all the mechanical elements should be obtained first; in other words, the existence of conventional transfer matrices is a prerequisite for the application of the RTMM. Thus, it seems that the RTMM is incapable of performing the dynamics analysis of linear multibody systems with indeterminate in-span conditions due to the nonexistence of the corresponding conventional transfer matrices. Observe that, for any state variables with indeterminate input–output relationships, the complementary state variables (the complementary state variable of a displacement is the corresponding internal force and vice versa) are identically equal to zero, and that the dimension of the Riccati transfer equation is only half of that of the conventional transfer equation. It reveals that the Riccati transfer equations for the connection points associated with indeterminate in-span conditions can be formulated directly, and that there is no need to rely on the conventional transfer equation. Two numerical examples are simulated and the computational results are compared with those obtained by the finite element method, which verifies the proposed method.


Author(s):  
Qinbo Zhou ◽  
Xiaoting Rui

The obtaining of the overall transfer equation of a linear controlled multibody system is one of the key problems when the transfer matrix method for multibody systems (MSTMM) is applied to study controlled multibody dynamics. This paper applies the theory of matrix signal flow graphs (MSFG) to MSTMM. The transfer matrix signal flow graphs (TMSFG) of a mechanical element and a control subsystem can be drawn according to their transfer equations. By merging the nodes corresponding to the same state vector, the TMSFG of the entire system can be obtained. Eventually, the overall transfer equation of the linear controlled system can be systematically obtained according to the reduction rules of MSFG. An example is taken to sketch the idea and the simulation results are compared with other ordinary dynamics methods to validate the proposed method. This paper lays a foundation for automatic deduction of overall transfer equations of linear controlled multibody systems.


Author(s):  
Lina Zhang ◽  
Xiaoting Rui ◽  
Jianshu Zhang ◽  
Junjie Gu ◽  
Huaqing Zheng ◽  
...  

Abstract The conventional transfer matrix method for multibody systems (MSTMM) with closed-loops (CLs) has superiority of avoiding the global dynamics equations. However, it requires a transfer equation to link Multiple-Input Single-Output (MISO) rigid body with multi-hinge subset and supplement equations caused CLs. In order to simplify the deduction processing and improve the numerical stability, the Riccati transformation is introduced and the Riccati transfer matrix method for multibody systems (RMSTMM) with CL is proposed. In a system with CLs, each CL is cut off at the connection point, and the new unknowns generated at the cut-off point are introduced into the Riccati recurrence relation. The numerical results of the conventional MSTMM and the RMSTMM are compared, and the reliability of the RMSTMM is verified. Meanwhile, the constrained Jacobian matrix is used to eliminate the non-working reactions of the system. The variations of the constraint violation error are compared to validate necessarily of constraints.


Author(s):  
Tianxiong Tu ◽  
Guoping Wang ◽  
Xiaoting Rui ◽  
Jianshu Zhang ◽  
Xiangzhen Zhou

Rui method, namely the transfer matrix method for multibody systems (MSTMM) is a new and efficient method for multibody system dynamics (MSD) for its features as follows: without global dynamics equations of the system, high programming, low order of system matrix and high computational speed. Riccati transfer matrix method for multibody systems was developed by introducing Riccati transformation in MSTMM, for improving numerical stability of MSTMM. In this paper, based on Riccati MSTMM, applying the thought of direct differentiation method, by differentiation of Riccati transfer equations of rigid bodies and joints, generalized acceleration and its differentiation can be obtained. Combined with Backward Euler algorithm, implicit algorithm for Riccati MSTMM is proposed in this paper. The formulation and computing procedure of the method are presented. The numerical examples show that results obtained by first order accurate implicit algorithm proposed in the paper and the fourth order accurate Runge-Kutta method have good agreement, which indicates that this implicit method is more numerical stability than explicit algorithm with the same order accurate. The implicit algorithm for Riccati MSTMM can be used for improving the computational accuracy of multibody system dynamics.


Sign in / Sign up

Export Citation Format

Share Document