| Summary: | 碩士 === 中原大學 === 機械工程研究所 === 106 === Restriction or constant flow is utilized to compensate recess pressures of a hydrostatic bearing which can be highest load capacity than any other type bearing, but not stiffness. Like to other type bearings supporting the worktable that is displaced continuously with changing load. This study concerns flow rate control for constant thickness of hydrostatic film. As long as film thickness is controlled to be constant, worktable can be kept stable without displacement when load is changing.
The objective of this study is to maintain the constant film thickness of the oil film thickness of the hydrostatic bearing closed table, and to discuss the pressure control of the pressure valve to solve the problem of film thickness change caused by the change of load, and to supply oil on the hydrostatic bearing closed workbench. The pressure, displacement meter signal and flow signal are used as feedback signals, and the control circuit is established by the pressure compensation control hydrostatic pressure table. In this paper, the pressure sensor and the displacement sensor are used to feedback the signal, and then the pressure difference and the displacement difference are calculated for pressure compensation to achieve the target value of the constant film thickness.
In the constant film thickness pressure compensation system, it is necessary to use the neural network training and the PID controller system to achieve a transient and fast convergence, and the steady state can achieve a precise and stable compensation system. The current compensation method of hydrostatic bearing pressure is made, and the calculation of the control law is made. The mathematical model is proposed and verified by the film thickness control of the pressure control valve.
The steady-state accuracy in few sub-micrometers can be achieved by Neueal Network and PID controller using the algorithm of this study, which have been verified by numerical simulation and measurement in experiments.
|
|---|
