Search results

A neural network is employed to analyze axial piston pump of hydrostatic circular recessed bearing. Owing to complexity of the system, the neural network is used to predict the bearing parameters of the experimental system. The system mainly consists of cylinder block, piston, slipper, ball‐joint and swash plate. The neural model of the system has three layers, which are input layer with one neuron, hidden layer with ten neurons and output layer with three neurons. It can be outlined from the results for both approaches neural network could be modeled bearing systems in real time applications.

The main objective of this study was to obtain the flow restricting capacity by determining their flow coefficients and to investigate the unsteady flow with low Reynolds number in the flow‐restricting devices such as orifices and capillary tubes having small diameters.

There is an enormous literature on the flow of Newtonian fluids through capillaries and orifices particularly in many application fields of the mechanical and chemical engineering. But most of the experimental results in literature are given for steady flows at moderate and high Reynolds numbers (Re>500). In this study, the unsteady flow at low Reynolds number (10<Re<650) through flow‐restricting devices such as orifices and capillary tubes having very small diameters between 0.35 and 0.70 mm were experimentally investigated.

The capillary tubes have much more capillarity property with respect to equal diameter orifices. Increasing the ratio of capillary tube length to tube diameter and decreasing the ratio of orifice diameter to pipe diameter before orifice increase the throttling or restricting property of the orifices and the capillary tubes. The orifices can be preferred to the capillary tubes having the same diameter at the same system pressure for the hydraulic systems or circuits requiring small velocity variations. The capillary tubes provide higher pressure losses and they can be also used as hydraulic accumulators in hydraulic control devices to attenuate flow‐induced vibrations because of their large pressure coefficients. An important feature of the results obtained for capillary tubes and small orifices is that as the d/D for orifices increases and the L/d reduces for capillary tubes, higher values C are obtained and the transition from viscous to inertia‐controlled flow appears to take place at lower Reynolds numbers. This may be explained by the fact that for small orifices with high d/D ratios and for capillary tubes with small L/d ratios, the losses due to viscous shear are small. Another important feature of the results is that the least variations in C for small orifices and the higher variations in C for capillary tubes occur when the d/D and L/d ratios are smallest. This has favourable implications in hydraulic control devices since a constant value for the C may be assumed even at relatively low values of Re.

To the authors' knowledge, there is not enough information in the literature about the flow coefficients of unsteady flows through capillary tubes and small orifices at low Reynolds numbers. This paper fulfils this gap.

This paper presents an investigation for analyzing the efficiency of axial piston pumps in a variety conditions using a proposed neural network. Since slippers affect the performance of the system considerably, the effects of surface roughness on lubrication have been studied in slippers with varying hydrostatic bearing areas and surface roughness. The neural network structure is very suitable for this kind of system. The network is capable of predicting the leakage oil quantity of the experimental system. The network has parallel structure and fast learning capacity. It is also easy to see from the experimental results that the leakage oil quantity was caused by surface roughness, orifice diameter and the size of hydrostatic bearing area, loading pressure and the number of rotations. It can be outlined from the results for both approaches, neural network could be modeled slipper bearing systems in real time applications.