Search

This paper deals with the prediction of flank wear of the cutting tool insert and surface roughness of the machined surface in hard turning using artificial intelligence technique. During machining of hard materials, the flank wear and surface roughness are favoured by both machining parameters and geometrical parameters. Prediction of output responses is much more important since surface roughness produced depends on the wear occurring at the flank face of the cutting tool insert. Based on the chosen input control parameters an L18 orthogonal array is chosen for seven input parameters varied through three levels to perform experiments using Taguchi's design of experiments. An artificial neural network (ANN) model is developed to predict the flank wear and surface roughness, which is compared with the predicted values from the developed empirical equations using multiple linear regression models. The results obtained from the ANN model shows a good prediction when comparing with the regression models showing the superiority of the generated neural network model. [ABSTRACT FROM AUTHOR]

In this paper, a novel dynamic compensatory strategy based on radial basis function (RBF) neural network robust H∞ controller is put forward for improving trajectory tracking accuracy of robot manipulators on account of modelling errors and external disturbances. The proposed strategy is effective to compensate the computed torque controller relied on precise dynamic model of robot, consisting of a RBF neural network, a variable structure controller and a robust H∞ controller. Furthermore, a velocity feed-forward with proposed method is employed to achieve higher trajectory tracking accuracy of end-effector and avoid the consideration of torque interface. The parameters of uncertain dynamics model are estimated by RBF neural network, while a variable structure controller is employed to make up the approximation errors of neural network. With the help of robust H∞ controller, effect of external disturbances will be attenuated and the robust tracking performance is achieved. Based on Lyapunov stability theorem, it is shown that the proposed controller can guarantee the stability of the control system and the desired performance of robotic system. The simulations and experiments on a multi-robot cutting manipulator indicate that the proposed compensatory strategy for industrial robot manipulator is superior to traditional computed torque control method in consideration of uncertainties. [ABSTRACT FROM AUTHOR]

Considering the modification of teeth, misalignment error and machining error, the paper proposes an approach for computerised simulation of double helical gears meshing. Algorithms for the tooth contact analysis (TCA) computer programs are presented. By using this program, the path of surface contact and transmission errors are obtained. Finally, an experiment that illustrates the proposed theory is provided. The results indicate that the TCA method of double helical gears is reliable. [ABSTRACT FROM AUTHOR]

Vacuum glazing is an innovative transparent thermal insulator that has applications in high performance windows. Two sheets of glass are separated by an evacuated space maintained by a regular array of support pillars, and the structure is sealed around the edges with solder glass or a metal bond. The evacuated internal gap reduces the heat transfer across the glazing by eliminating air conduction and convection between the glass sheets. This review paper traces the development of vacuum glazing from the initial proposal 100 years ago, to the fi rst reported manufacture in 1989 by the University of Sydney, the subsequent commercialisation of that design in 1996 and recent advances. The various design challenges are discussed along with the innovative ways they were successfully addressed by two research groups. Several key non-technical factors that have been equally important to the commercialisation and acceptance of this new technology are also presented. The future prospects for vacuum glazing are discussed in the context of the growing interest in energy effi cient buildings and higher standards for the thermal performance of windows. [ABSTRACT FROM AUTHOR]

In this paper, the process of bimetallic backward rod extrusion through a conical punch, by means of upper bound method and fi nite element method is investigated. An admissible velocity fi eld is developed and then by calculating the internal, shear and frictional powers, the extrusion force is estimated. The extrusion process is also simulated by using the fi nite element code, ABAQUS. Analysis and simulations are done for two types of bimetallic rods: aluminium as core, copper as sleeve (Al-Cu) and copper as core, aluminium as sleeve (Cu-Al). The extrusion force from the upper bound method is compared with the fi nite element results. This comparison shows that the upper bound predictions are in good agreement with the fi nite element results. The results also show that, the extrusion force in the case of Cu-Al rod is smaller than Al-Cu rod and in both types of bimetallic rods, the aluminium leaves the deformation zone sooner than the copper. Finally the effects of various extrusion parameters, such as the friction factor, reduction in area and semi-punch angle upon the extrusion force are investigated and the optimum semi-punch angle is determined. [ABSTRACT FROM AUTHOR]

Stiff ordinary differential equations (ODEs) are present in engineering, mathematics, and sciences. Identifying them for effective simulation is imperative. This paper considers only linear initial value problems and brings to light the fact that stiffness ratio or coeffi cient of a suspected stiff dynamic system can be elusive as regards the phenomenon stiffness. Though, it gives the insight suggesting stiffness when the value is up to 1000 but is not necessarily so in all ODEs. Neither does a value less than 1000 imply non-stiffness. Here we put forward that a stiff ODE among other considerations should also require only a stiff solver to easily approximate its solution, otherwise it is non-stiff and experimenting with different solvers is invaluable. Since no single numerical method of solving a model suffi ces for all systems, choice of a suitable solver for simulation is of utmost important. Hence, experimenting with all the variable-step solvers in Matlab ODE suit for every problem is invaluable. Results point to the fact that stiffness coeffi cient of about 1000 does not always suggest that an ODE is stiff nor does a value less than 1000 suggest non-stiff. Also, Matlab and Maple results were compared graphically with no disparity. [ABSTRACT FROM AUTHOR]