Sensitivity analysis is applied to an investigation of the influence of parameters on the dynamics of a variable displacement axial piston pump. Since an exact mathematical model of piston pump is complex and highly non-linear, the investigation of its dynamic characteristics for the variation of the system purameters by changing separate parameter values in sequence becomes very inefective. In this paper a parameter sensitivity analysis was employed to analyse the eflects of the system parameters systematically. The analysis was done on the exact non-linear dynamic model of the pump system which is represented by fourth-order dynamics. Based upon the simulation results, the degree of influence ofeuch parameter and the justification of order reduction are discussed in some detail.

The load-sensing hydraulic system is an energy saving hydraulic system which improves the efficiency of transmitting power from the pump to the load. However, its stability characteristics deteriorate critically due to the addition of the load-sensing mechanism, compared with those of the conventional system. In this pfi!per, a non-linear mathematical model of the load-sensing hydraulic system is formulated, taking into consideration the dynamics of the load-sensing pump. Based upon linearization of this model for various operating conditions, the stability analysis has been madt using the Routh-Hurwitz stability criterion. The results of the theoretical stability analysis were assured through experiments. Both results show that stability is critical to the choice of system parameters such as the setting pressure of the pump compensator and the load inertia. 3/88 0 IMechE 1988 0263-7138/88 $2.00 + .05

This paper introduces a new analytical solution to predict transient temperature distributions in a finite thickness plate during laser surface hardening. This analytical solution was obtained by solving a transient three-dimensional heat conduction equation with convection boundary conditions at the surfaces of the workpiece. To calculate the temperature field numerically in laser surface hardening processes, laser beam absorptivity, one of the most important parameters, should first be determined. It was extremely difficult to find an accurate value for laser beam absorptivity owing to the use of coating material on the workpiece surfaces, essentially to improve the efficiency of laser beam power. Therefore, in this paper, absorptivities were measured experimentally under various hardening conditions, including variations in coating thickness, laser beam power and beam travel velocity. Thereafter, to prove the validity of the model, a series of laser surface hardening experiments was performed on medium-carbon steel under various hardening conditions. The hardened layer of the etched cross-sectional view was examined and compared with the isothermal lines predicted by the proposed analytical model, in which the values of the experimentally measured absorptivity were employed. The results showed that the solution predicted well the transient temperature distribution of finite plates with satisfactory accuracy. Also, owing to the simplicity of the solution method, the analytical model developed may be easily implemented for simulation work for analysis and prediction of laser surface hardening processes under various hardening conditions.

Stereolithography has attracted more attention due to better part build accuracy than other rapid prototyping technologies. However, this build method still limits wider applications due to the unsatisfactory level of dimensional accuracy that remains with the current technology. To improve accuracy and reduce part distortion, understanding the physics involved in the relationship between the operating input parameters and the part dimensional accuracy is prerequisite. In this paper, this causality is identi®ed through a process model obtained via an arti®cial neural network based upon 140 actual build parts. The network is so constructed that it relates the process input parameters to part dimensional accuracy. The neural network model is found to predict the eVects of the input parameters on the accuracy with reasonable accuracy. The prediction performance is discussed in detail for various process parameter ranges.

Non-linear characteristics and uncertainty in manipulator dynamics caused by payload effects are major hurdles in controller design. To overcome such hurdles the authors have introduced an automatic balancing concept which has been proved to reduce the non-linear complexity in manipulator dynamics as well as to remove gravity loading. This paper examines the characteristic features of balanced manipulator dynamics in more detail and presents an efficient control algorithm suitable for the dynamics. Since the dynamics of a balanced manipulator are characterized by partially configuration-independent inertial properties, the present algorithm adopts two different control concepts ‘the computed torque control’ for the joint having coupled, configuration-dependent inertia and ‘an optimal constant feedback control’ for the joints having configuration-independent inertia. To evaluate the proposed control algorithm, simulation studies were made over a wide range of manipulator speeds and payloads. Based upon the simulation results, the efficiency of the controller is discussed in detail.