Motion equations in redundant coordinates with application to inverse dynamics of constrained mechanical systems

The basis for any model-based control of dynamical systems is a numerically efficient formulation of the motion equations, preferably expressed in terms of a minimal set of independent coordinates. To this end the coordinates of a constrained system are commonly split into a set of dependent and independent ones. The drawback of such coordinate partitioning is that the splitting is not globally valid since an atlas of local charts is required to globally parameterize the configuration space. Therefore different formulations in redundant coordinates have been proposed. They usually involve the inverse of the mass matrix and are computationally rather complex. In this paper an efficient formulation of the motion equations in redundant coordinates is presented for general non-holonomic systems that is valid in any regular configuration. This gives rise to a globally valid system of redundant differential equations. It is tailored for solving the inverse dynamics problem, and an explicit inverse dynamics solution is presented for general full-actuated systems. Moreover, the proposed formulation gives rise to a non-redundant system of motion equations for non-redundantly full-actuated systems that do not exhibit input singularities.     

A smoothed particle hydrodynamics-based fluid model with a spatially dependent viscosity: application to flow of a suspension with a non-Newtonian fluid matrix

A smoothed particle hydrodynamics approach is utilized to model a non-Newtonian fluid with a spatially varying viscosity. In the limit of constant viscosity, this approach recovers an earlier model for Newtonian fluids of Español and Revenga (Phys Rev E 67:026705, 2003). Results are compared with numerical solutions of the general Navier–Strokes equation using the “regularized” Bingham model of Papanastasiou (J Rheol 31:385–404, 1987) that has a shear-rate-dependent viscosity. As an application of this model, the effect of having a non-Newtonian fluid matrix, with a shear-rate-dependent viscosity in a moderately dense suspension, is examined. Simulation results are then compared with experiments on mono-size silica spheres in a shear-thinning fluid and for sand in a calcium carbonate paste. Excellent agreement is found between simulation and experiment. These results indicate that measurements of the shear viscosity of simple shear-rate-dependent non-Newtonian fluids may be used in simulation to predict the viscosity of concentrated suspensions having the same matrix fluid.     

Stochastic response determination of structural systems modeled via dependent coordinates: a frequency domain treatment based on generalized modal analysis

Generalized independent coordinates are typically utilized within an analytical dynamics framework to model the motion of structural and mechanical engineering systems. Nevertheless, for complex systems, such as multi-body structures, an explicit formulation of the equations of motion by utilizing generalized, independent, coordinates can be a daunting task. In this regard, employing a set of redundant coordinates can facilitate the formulation of the governing dynamics equations. In this setting, however, standard response analysis techniques cannot be applied in a straightforward manner. For instance, defining and determining a transfer function within a frequency domain response analysis framework is challenging due to the presence of singular matrices, and thus, the machinery of generalized matrix inverses needs to be employed. An efficient frequency domain response analysis methodology for structural dynamical systems modeled via dependent coordinates is developed herein. This is done by resorting to the Moore–Penrose generalized matrix inverse in conjunction with a recently proposed extended modal analysis treatment. It is shown that not only the formulation is efficient in drastically reducing the computational cost when compared to a straightforward numerical evaluation of the involved generalized inverses, but also facilitates the derivation and implementation of the celebrated random vibration input–output frequency domain relationship between the excitation and the response power spectrum matrices. The validity of the methodology is demonstrated by considering a multi-degree-of-freedom shear type structure and a multi-body structural system as numerical examples.

     

Stability analysis of relative equilibria of mechanical systems with cyclic coordinates: a direct approach

Nonlinear stability of relative equilibria of mechanical systems has been investigated during the past two decades by notable authors and has resulted in the so-called energy momentum method. Although it has numerous important engineering applications, this theory involves subtle mathematical methods such as group theory with which engineers usually are not familiar. This paper develops a simple and natural approach to the problem for the case of cyclic coordinates in the Lagrangian since many practical examples can be easily formulated in terms of cyclic coordinates. Referring to standard algebraic operations, a stability criterion for relative equilibria is derived. As a computational benefit the presented approach does not require knowledge of a system's complete kinetic energy, either for formulating steady-state equations or for checking stability. The application of the method, which is closely related to Routh's method, will be demonstrated using the example of a dumbell satellite.     

一类广义连续-离散系统量测丢失情况下鲁棒滤波算法

有噪声的非线性广义连续-离散系统的状态估计问题是目前一个较新的研究领域。量测丢失可能会导致系统动态模型的状态估计值与动态方程真实值产生较大偏差。为解决该问题,针对一类量测丢失情况下的范数有界非线性广义连续-离散系统,提出一种基于鲁棒扩展卡尔曼滤波(REKF)算法的状态估计方法。首先,给出参数并使用欧拉离散化方法将理想化的广义连续-离散系统转化为非奇异离散系统,但这样的处理方式导致转换得到的非奇异离散系统动态模型中存在新增不确定项。针对该问题,提出最优上界以保证卡尔曼滤波误差协方差矩阵收敛。其次,针对转化得到的非奇异离散系统,提出基于该优化上界的扩展卡尔曼滤波方法,用于在量测丢失时对系统的动态模型量进行观测。最后,仿真算例验证了该方法的有效性。     

Robust output feedback control of time-delay nonlinear systems with dead-zone input and application to chemical reactor system

Nonlinear Dynamics - This paper investigates the problem of robust output feedback control for a class of time-delay nonlinear systems with unknown continuous time varying output function. Unlike...     

Cosserat interphase models for elasticity with application to the interphase bonding a spherical inclusion to an infinite matrix

Interphases are often modeled as interfaces with zero thickness using jump conditions that can be developed based on approximate shell or membrane models which are valid for specific limited ranges of the elastic material parameters. For a two-dimensional problem it has been shown (Rubin and Benveniste, 2004) that the Cosserat model of a finite thickness interphase is a unified model that is accurate over the full range of elastic parameters. In contrast, many other interphase models are valid for only limited ranges of the elastic parameters. In this paper, the accuracy of different Cosserat models of a finite thickness interphase that connects a spherical inclusion to an infinite matrix is examined. Specifically, four Cosserat interphase models are considered: a general shell (GS)$(\mathit{GS})$, a membrane-like shell (MS)$(\mathit{MS})$, a simple shell (SS)$(\mathit{SS})$ and a generalized membrane (GM)$(\mathit{GM})$. The models (GS)$(\mathit{GS})$ and (MS)$(\mathit{MS})$ both satisfy restrictions on the strain energy function of the interphase that ensure exact solutions for all homogeneous three-dimensional deformations, while the other models (SS)$(\mathit{SS})$ and (GM)$(\mathit{GM})$ do not satisfy these restrictions. The importance of these restrictions is examined for the three-dimensional inhomogeneous inclusion problem being considered. This is the first test of the accuracy of an elastic interphase model for a spherical interphase.     

Observer-based controller for discrete-time systems: a state dependent Riccati equation approach

In this paper, an observer-based controller for discrete-time nonlinear dynamical systems is proposed. After transforming the nonlinear system to a linear structure having state-dependent coefficient matrices (SDC), a recursive regularized least-square (RLS) state estimator is developed. The observed states are then used to generate either a constrained or unconstrained state feedback controller using the state dependent Riccati equation (SDRE) approach. The stability of the observer-based control system is rigorously analyzed in a theoretical frame work. Applications to different numerical examples as well as to a practical case study demonstrate the effectiveness of the proposed procedure.     

Robust adaptive fractional-order observer for a class of fractional-order nonlinear systems with unknown parameters

Nonlinear Dynamics - This paper investigates the parameter and state estimation problems for a class of fractional-order nonlinear systems subject to the perturbation on the observer gain. The...     

Numerical study of the noninertial systems: application to train coupler systems

Car coupler forces have a significant effect on the longitudinal train dynamics and stability. Because the coupler inertia is relatively small in comparison with the car inertia; the high stiffness associated with the coupler components can lead to high frequencies that adversely impact the computational efficiency of train models. The objective of this investigation is to study the effect of the coupler inertia on the train dynamics and on the computational efficiency as measured by the simulation time. To this end, two different models are developed for the car couplers; one model, called the inertial coupler model, includes the effect of the coupler inertia, while in the other model, called the noninertial model, the effect of the coupler inertia is neglected. Both inertial and noninertial coupler models used in this investigation are assumed to have the same coupler kinematic degrees of freedom that capture geometric nonlinearities and allow for the relative translation of the draft gears and end of car cushioning (EOC) devices as well as the relative rotation of the coupler shank. In both models, the coupler kinematic equations are expressed in terms of the car body and coupler coordinates. Both the inertial and noninertial models used in this study lead to a system of differential and algebraic equations that are solved simultaneously in order to determine the coordinates of the cars and couplers. In the case of the inertial model, the coupler kinematics is described using the absolute Cartesian coordinates, and the algebraic equations describe the kinematic constraints imposed on the motion of the system. In this case of the inertial model, the constraint equations are satisfied at the position, velocity, and acceleration levels. In the case of the noninertial model, the equations of motion are developed using the relative joint coordinates, thereby eliminating systematically the algebraic equations that represent the kinematic constraints. A quasistatic force analysis is used to determine a set of coupler nonlinear force algebraic equations for a given car configuration. These nonlinear force algebraic equations are solved iteratively to determine the coupler noninertial coordinates which enter into the formulation of the equations of motion of the train cars. The results obtained in this study showed that the neglect of the coupler inertia eliminates high frequency oscillations that can negatively impact the computational efficiency. The effect of these high frequencies that are attributed to the coupler inertia on the simulation time is examined using frequency and eigenvalue analyses. While the neglect of the coupler inertia leads, as demonstrated in this investigation, to a much more efficient model, the results obtained using the inertial and noninertial coupler models show good agreement, demonstrating that the coupler inertia can be neglected without having an adverse effect on the accuracy of the solution.     

A minimum theorem for cyclic load in excess of shakedown,with application to the evaluation of a ratchet limit

Although the shakedown theorems for perfect plasticity have been known since Koiter's 1960 review paper, extensions of the theory to situations where ratchetting or reverse plasticity occurs in excess of shakedown have not appeared in the literature. In this paper a generalisation of the upper bound theorem is derived which reduces to the upper bound shakedown theorem in the limiting case when the load point approaches the shakedown boundary. The new theory is used to develop a method for identifying the ratchet limit for a class of loading histories through the sequential minimisation of two functionals. A programming method, based on the Elastic Compensation method for shakedown is then derived and convergence proven. Numerical examples of the application of the method to practical problems are discussed by us in an accompanying paper.     

Robust adaptive synchronization for a class of chaotic systems with actuator failures and nonlinear uncertainty

This paper is concerned with the robust adaptive synchronization problem for a class of chaotic systems with actuator failures and unknown nonlinear uncertainty. Combining adaptive method and linear matrix inequality (LMI) technique, a novel type of robust adaptive reliable synchronization controller is proposed, which can eliminate the effect of actuator fault and nonlinear uncertainty on systems. After solving a set of LMIs, synchronization error between the master chaotic and slave chaotic systems can converge asymptotically to zero. Finally, illustrate examples about chaotic Chua’s circuit system and Lorenz systems are provided to demonstrate the effectiveness and applicability of the proposed design method.     

Robust identification for fault detection in the presence of non-Gaussian noises: application to hydraulic servo drives

Nonlinear Dynamics - Intensive research in the field of mathematical modeling of hydraulic servo systems has shown that their mathematical models have many important details which cannot be...     

Robust optimal control for constrained uncertain switched systems subjected to input saturation: the adaptive event-triggered case

Nonlinear Dynamics - This paper investigates the adaptive event-triggered robust optimal control approach for discrete-time switched systems subjected to input saturations and exogenous...     

An analytical model of a matrix heat exchanger with longitudinal heat conduction in the matrix in application for the exchanger dynamics research

The paper presents an analytical method for the solution of the problem of a fixed matrix heat exchanger with axial heat conduction within the matrix. The small parameter method and Laplace transform have been applied. A general solution has been obtained for the unsteady state in the form of function series, using single and double convolutions of functions, as well as a particular solution for both the uniform and non-uniform initial temperature of the matrix and for an arbitrary function of the fluid temperature at the inlet. Particular solutions have been used in the study of the matrix dynamics in determining dynamic characteristics for the standard input signals in the form of: Dirac pulse, Heaviside function and the function of sinusoidal variable temperature of the fluid at the inlet. The results obtained both illustrate and enable the assessment of the effect of axial heat conduction in the matrix on the dynamic properties of the heat exchanger.

---

Robust adaptive intelligent sliding model control for a class of uncertain chaotic systems with unknown time-delay

In this paper, a robust adaptive intelligent sliding model control (RAISMC) scheme for a class of uncertain chaotic systems with unknown time-delay is proposed. A sliding surface dynamic is appropriately constructed to guarantee the reachability of the specified sliding surface. Within this scheme, neuro-fuzzy network (NFN) is utilized to approximate the unknown continuous function. The robust controller is an adaptive controller used to dispel the unknown uncertainty and approximation errors. The adaptive parameters of the control system are tuned on-line by the derived adaptive laws based on a Lyapunov stability analysis. Using appropriate Lyapunov–Krasovskii (L–K) functional in the Lyapunov function candidate, the uncertainty caused by unknown time delay is compensated and the global asymptotic stability of the error dynamics system in the specified switching surface is accomplished. Finally, the proposed RAISMC system is applied to control a Hopfield neural network, Cellular neural networks, Rössler system, and to achieve synchronization between the Chen system with two time delays with Rössler system without time delay. The results are representative of outperformance of the proposed method in all cases.     

On dynamics and stability of continuous systems subjected to a distributed moving load

Summary The paper deals with an analysis of different models of continuous systems subjected to a load distributed over a given length and moving at a constant velocity.The general discussion concerns a beam resting on a viscoelastic semi-space. The motion of the body is described by polynomial differential operators. With body forces being disregarded, the motion of the beam lying on a viscoelastic foundation is discussed with either taking into account the effect of shear deflection and the inertia of rotation or with neglecting them. Consideration is also given to the cases of relative motion of two continuous systems and the stability of their interaction. The above cases represent the models of a number of mechanical systems applied in e.g. modern transportation facilities, technology of bonding of layer materials, etc.The analysis is substantially simplified because the equations of motion and the boundary conditions are written in a moving coordinate system related with the load and only stationary solutions are considered. With such an approach, the solutions depend only on the transformed space variable, while the velocity of the load appears as one of the parameters of the system. Interesting conclusions are drawn from the obtained solutions and numerical calculations.

---

Zur Dynamik und Stabilität von kontinuierlichen Systemen unter verteilten bewegten Lasten

Übersicht Der vorliegende Beitrag behandelt die Analyse von kontinuierlichen Systemen, die verteilten bewegten Lasten ausgesetzt sind. Es wird zunächst ein Balken auf viskoelastischem Halbraum behandelt. Die Bewegung des Körpers wird beschrieben durch Differentialoperatoren in Polynomform, wobei der Einfluß von Schubverformung und Drehträgheit diskutiert wird. Der Fall der Relativbewegung von zwei kontinuierlichen Systemen und ihre Stabilität wird ebenfalls diskutiert. Die untersuchten Fälle stellen Modelle zahlreicher mechanischer Systeme dar, wie zum Beispiel moderne Schnelltransportsysteme, Fügevorrichtungen von Schichtmaterialien, usw.Die Analyse der Probleme wird wesentlich dadurch vereinfacht, daß die Bewegungsgleichungen und Randbedingungen in einem mitbewegten Koordinatensystem formuliert werden und nur stationäre Lösungen betrachtet werden. Auf diese Weise hängen die Lösungen nur von der transformierten Ortsvariablen ab; die Geschwindigkeit der Belastung tritt als ein Parameter des Systems auf.Aufschlußreiche Schlußfolgerungen werden anhand der erhaltenen Lösungen und numerischen Berechnungen dargestellt.

---

---

This paper was presented at the Second Symposium on Inelastic Solids and Structures, Bad Honnef, September 1981