The Research On Pneumatic Constant Pressure Output System Accuracy And Stability

Pneumatic technology, with its many advantages are widely used in various fields, especially applied to fast, clean, flexible output required. In the field of mechanical processing, pneumatic technology used in grinding and polishing surface, make the surface roughness of the workpiece to arrive at the expected value.And the compressibility of the gas and influenced by temperature, make the pneumatic system of the natural frequency is low, poor linearity, it is difficult to used for high precision control.At the same time under the influence of friction in the cylinder, cylinder will have serious when running at low speed crawl, it will seriously affect the performance of the system.This article will revolve around the lens polishing machine pneumatic stability constant force output system is analyzed and discussed.In order to make the pneumatic proportional system output force under interference can reach high precision, in the low speed range does not produce crawl and improve the stiffness of pneumatic system, this article will be to improve the cylinder piston structure, but also by using BP neural network PID control as the control strategy of pneumatic servo control system, to adapt to the nonlinear nature of pneumatic system.Through these two measures, can effectively improve the performance of the system.The article firstly establish pneumatic servo system mathematical model,the reason of creating mathematical model is to acquire inner relationship of the system. Mathematical models can obtaine relationship between input and output of the system, and the influence of the system output of the internal parameters, in order to analyze the system and adjusting system parameters to achieve the optimal state of the system.To meet accuracy and stability of the polishing machine pneumatic loading system, the paper will improve cylinder partial structure. It will adopt the gas film supporting in the cylinder piston and cylinder contact area, to reduce or eliminate the friction of the cylinder. Meanwhile, in order to compensate for the shortcomings of low stiffness gas in the cylinder,to use moderate stiffness spring in the two chambers of the cylinder, it can compensate shortcomings of the poor stiffness gas.Finally established AMESim simulation model, and then to some factors which influence the stability of the cylinder and output force are simulated, that influence the stability of the system output power relationship, to guide the design of the constant force output system.Gave an analysis of the basic knowledge of the neural network, then gives the principle of BP neural network PID control and steps. Write a BP neural network PID control program, and run in MATLAB to get step and sinusoidal input response.By the simulation results shows that neural network has fast response, tracking ability is strong, can be applied to constant pressure output on the pneumatic system.