An automatic method for controlling the centre of gravity of a model

In many applications the location of the centre of gravity of a mechanical part is an important factor that a designer must consider. If it is not in a desired location, a part might not work properly, e.g. unbalanced force might be generated in a rotational part. After a part is modeled, its centre of gravity cannot be altered unless its external shape or internal mass distribution is changed. However, the external shape is usually constrained by other design considerations. In this paper, an algorithm is proposed for controlling the centre of gravity of a hollowed part. Using this algorithm, the location of the centre of gravity of a part is controlled by changing its internal mass distribution.     

A time aggregation approach to Markov decision processes

We propose a time aggregation approach for the solution of infinite horizon average cost Markov decision processes via policy iteration. In this approach, policy update is only carried out when the process visits a subset of the state space. As in state aggregation, this approach leads to a reduced state space, which may lead to a substantial reduction in computational and storage requirements, especially for problems with certain structural properties. However, in contrast to state aggregation, which generally results in an approximate model due to the loss of Markov property, time aggregation suffers no loss of accuracy, because the Markov property is preserved. Single sample path-based estimation algorithms are developed that allow the time aggregation approach to be implemented on-line for practical systems. Some numerical and simulation examples are presented to illustrate the ideas and potential computational savings.     

Inversion of confluent Vandermonde matrices

This paper is concerned with the inversion of confluent Vandermonde matrices. A novel and simple recursive algorithm for inverting confluent Vandermonde matrices is presented. The algorithm is suitable for classroom use in both numerical as well as symbolic computation. Examples are included to illustrate the proposed algorithm.     

Prediction of residual colour in reactive dyeing effluent

In this study, the absorbance of the three-dye mixture effluent from reactive dyeing has been used to predict the residual colour value of the dyeing effluent. The colour values of the dyeing effluent have been predicted in terms of ADMI colour values. The prediction accuracy was found to be satisfactory when some mathematical approaches, such as linear and non-linear regression, were used.     

Study of the surface layer in cBN growth by PVD techniques

A sp² bonded boron nitride (sp² BN) surface layer is sometimes observed on top of cubic boron nitride (cBN) grains for BN films deposited using ion-assisted PVD techniques. Understanding the formation of such surface layer gives a clue to the cBN growth mechanisms. In the current study, the microstructure and phase composition near the top surface of several BN films were investigated in order to clarify the formation mechanism of such a surface layer. All the films investigated were synthesized using ion-assisted PVD techniques, but with different deposition parameters. It was found that such a surface layer is not necessarily present in some of the cBN growth, and its presence depends on the bombardment of ion energy during deposition. The cBN growth mechanisms are discussed based on these observations.     

An anisotropic theory of elasticity for continuum damage mechanics

This paper presents the development of an anisotropic elastic damage theory. This is achieved by deriving a modified damage effect tensor M(D) for the effective stress equations capable of including the effect of anisotropic material damage. The modified tensor removes the restriction of a priori knowledge of the directions of principal stresses imposed by a damage effect tensor developed earlier and can now be made for general practical engineering applications of failure analysis. Reduction of the proposed tensor to a scalar for isotropic damage is shown to be possible when it is expressed not only in the principal directions but also in any arbitrary coordinate system, a necessary condition to verify the validity of the proposed tensor. Uniaxial tension and pure torsion are chosen to illustrate the application of the theory as well as associated damage variables that may be experimentally determined using laboratory size specimens. The measured damage variables confirm the presence of anisotropic damage from an initially isotropic material specimen and the magnitude is more pronounced at higher stresses and strains.

---

Résumé On présente un développement d'une théorie sur l'endommagement élastique anisotrope en déduisant un tenseur modifié décrivant l'effet de l'endommagement pour un système d'équations de contraintes effectives susceptible d'inclure l'effet d'un endommagement dans un matériau anisotrope. Le tenseur modifié supprime la restriction de la connaissance a priori des directions des contraintes principales imposées par un tenseur d'effet d'endommagement développé précédemment; il peut à présent entrer dans les applications pratiques en construction de l'analyse des ruptures.On montre qu'il est possible de réduire le tenseur proposé à une valeur scalaire dans le cas d'un dommage isotrope, dès lors qu'il est exprimé non seulement suivant les directions principales, mais dans un système de coordonnées arbitraires, ce qui est une condition nécessaire pour en vérifier la validité.On choisit une traction multiaxiale et une torsion pure pour illustrer l'application de la théorie ainsi que des variables d'endommagement associées, susceptibles d'être déterminées expérimentalement à l'aide d'éprouvettes de laboratoire.Les variables d'endommagement mesurées confirment la présence d'un dommage anisotrope dans le cas d'une éprouvette d'un matériau initialement isotrope; son amplitude est plus prononcée à des contraintes ou des déformations plus importantes.

     

Numerical simulation model of vibration responses of rectangular plates embedded with piezoelectric actuators

A numerical simulation model for random large amplitude vibration control of composite plate using piezoelectric material is presented. The H_∞ control design is employed to suppress the large amplitude vibrations of composites plates under random loading. The numerical simulation model is developed and based on the finite element method. The finite element governing equation includes fully coupled structural and electrical nodal degrees of freedom, and consider the von Karman large amplitude vibration. The modal reduction method using the structural modes is adopted to reduce the finite element equations into a set of modal equations with fewer degrees of freedom. The modal equations are then employed for controller design and time domain simulation. In the simulations without control, the value of the linear mode to the nonlinear deflection is quantified; and the minimum number of linear modes needed for accurate model is obtained. In the simulations with control, it is shown that the truncated modes, which are neglected in the control design, deteriorate the controller performance. Generally, the vibration reduction level is not monotonically increasing with the size of the piezoelectric actuator. The optimal piezoelectric actuator size depends on the excitation level. For higher excitation level, optimal actuator size is larger. The H_∞ controller based on the linear finite element formulation gives better vibration reduction for small amplitude vibration, but it still gives reasonable performance for large amplitude vibration provided that the piezoelectric actuator is big and powerful enough.     

Numerical simulation of aerosol collection in filters with staggered parallel rectangular fibres

 In this paper, filters with rectangular fibres arranged in a staggered and parallel array and placed transverse to the flow are studied numerically. A two- dimensional flow field is obtained by solving Navier–Stokes equations with the control volume method. Periodic boundary conditions are introduced in the calculation. In order to achieve higher accuracy, a second-order upwind scheme is adopted and a fine mesh is arranged near the fibre and the symmetrical plane of the flow field where large gradients in velocity are expected. Particle trajectories are calculated by solving the corresponding Lagrangian equation of motion to obtain the collection efficiency of a single rectangular fibre, in which positions of the approaching particles on the inlet plane of the flow field are randomly distributed according to the Monte-Carlo principle. The simulation considers all the important mechanisms of particle capture including interception, inertial impaction and Brownian motion. Effects of fibre aspect ratio, filter packing density, particulate size and Reynolds number on the collection efficiency are numerically determined. The volumetric packing density ranges from 0.4 to 4% and the particle diameter is from 0.01 μm to 2 μm. Reynolds number based on the height of computational domain varies from 20 to 100 and the aspect ratio is from 0.1 to 10. Simulations with and without Brownian motion are carried out for different Reynolds numbers, packing densities and aspect ratios and the results show that Brownian effects are significant for particles smaller than 1 μm. Received 25 May 2001     

Mechanism analysis of gate-induced drain leakage in off-state n-MOSFET

An analytical expression for both band-to-band and band-trap-band indirect tunnelings is used to study the gate-induced drain leakage (GIDL) current of MOSFETs measured before and after hot-carrier stress. The voltage and temperature dependence of GIDL are characterized. Both results show that interface traps situated near the midgap participate in the conduction of GIDL, and band-trap-band indirect tunneling could be the major mechanism. This is further supported by the fact that the percentage increase in GIDL induced by hot-carrier stress is about the same as the corresponding increase in interface-trap density. On the other hand, under low-field conditions, trap-assisted Poole–Frenkle emission dominates over tunneling for temperatures even well below room temperature.