Research on Vacuum Micro-clamp of Micro Manipulator Based on Fuzzy PD Control
Research on vacuum micro-clips of micro-manipulators based on fuzzy PD control CHEN Guo-liang, HUANG Xin-han, WANG Min (Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China) is an ideal strategy. A vacuum micro-clipping tool for the assembly of particles (100-300 m in diameter) was designed. A two-level control structure with A-Duc812 as the bottom controller and PC as the top controller is designed. The influence of the adhesion force of the microparticles and the stress conditions were analyzed, and the working pressure conditions of the microclips at different working stages were deduced. Considering the compressibility of air, the nonlinearity of pneumatic components and the pressure loss of micro-clips, a pressure controller based on fuzzy PD was designed to improve the ability of micro-grippers to pick up and release particles. Matlab simulation test results show the effectiveness of the controller. The micro-clip has the advantages of stable and reliable performance and compact structure, and can be suitably applied to assembly work for sub-millimeter particles.
1 vacuum microclip design can be ignored. Since the ball cannot be overcome by gravity, the external force can only be achieved with a small volume and mass for a small object. How to apply external force and how to determine and control the external force is a key issue.
1 Fund Project: National High-tech Research and Development Program (200 packs attack 84110) 01 The National Science Foundation Project studied the various effects of adhesion on release operations and proposed some solutions.
By analyzing the designed vacuum micro-clips, the main reason that the micro-clips can not complete the task of releasing the ball is that the micro-clips adopt the open-loop method to adjust the working pressure. The pneumatic system has the advantages of low damping, high compressibility, pipeline friction and other factors, which makes it difficult for the vacuum micro-gripper to achieve a satisfactory pressure regulation effect by only using open-loop proportional control. This paper redesigned the vacuum microclip and its pressure controller based on the microclip. The vacuum clamp consists of a vacuum unit and a control unit.
1.1 Vacuum Unit The vacuum unit consists of a gas source, a proportional pressure valve, a vacuum generator, a solenoid switch valve, a speed valve, a vacuum filter, a pressure gauge, and a pressure sensor (see ). The vacuum generator is the core device of the vacuum unit. The ZH07BS-06-06 type vacuum generator of SMC is used. The maximum vacuum degree is 88kPa and the working pressure range is 0.25~0.6MPa. The electric proportional pressure valve is selected as the core component of the pneumatic control system. Model 23BPD1, the proportional valve pressure control range of 0.05 ~ 0.9MPa. Vacuum micro-clip basic workflow is: pick up the ball, the solenoid valve 1 power, according to the Venturi principle, compressed air through the solenoid valve 1 and vacuum occurs Ejector, so that the vacuum adsorption port produces negative pressure adsorption ball; when the ball is released, the solenoid valve 1 is powered off, the vacuum generator stops working, the vacuum adsorption force quickly disappears, the solenoid valve 2 is powered on, and the positive pressure release gas is connected. On the way, the micro-clip creates a positive pressure difference to release the ball.
1.2 Control Unit The vacuum micro-clamp control unit adopts a two-stage control structure. Its upper layer is controlled by a PC, and is integrated with the other control units of the manipulator. The bottom layer control is implemented by the ADMC812 single chip microcomputer of AD Company. The ADMC812 communicates with the PC through the serial port for field data and control commands. The architectural diagram of the control unit.
The PC is at the heart of the control unit. The main functions of the upper PC control unit are: triggering and monitoring the work process of the underlying MCU unit, working pressure calibration, and control decisions. The PC control program based on the WINDOWS98 operating system adopts a programmable, self-calibrating, high-accuracy analog data acquisition system in the sense of 4YCf. The main functions of the bottom control unit are: on-off control of solenoid valves 1 and 2, pressure signal acquisition, AD, DA conversion, proportional pressure valve control, and LED display. The low-level unit control program is written by Franklin C51, adopts modular design, and can be divided into main program module, communication module, data acquisition module, LED display module, and control module according to the implemented functions.
2 Mechanical analysis The straw made of sodium carbonate or boron carbonate glass material was used as the adsorption element. For fine objects below submillimeters, the adhesion force exceeds gravity and plays a major role. Adhesion is mainly composed of three main forces: electrostatic adsorption force, van der Waals force and surface tension (di+d2), which is a simplified diameter, and di and d2 are the diameters of the two acting objects, respectively. The effect of electrostatic adsorption can be reduced by ionizing the surrounding environment or conducting static electricity on the surface.
Van der Waals force is the interaction force between the motion electrons, which can be calculated by: H is the Hamakter coefficient; 0 is the diameter of the ball; it is the effective distance between the ball and the contact surface. Rough surfaces can increase the effective distance z, thereby reducing van der Waals forces.
Surface tension is the most important surface force component. It is the force generated by the liquid film phenomenon between the contact surfaces of the object. Its size can be approximated by F(:ap=TtDY): D is the diameter of the liquid film; Y is Hydrophilicity coefficient, apparently hydrophobic surface of the object and increase air dryness can reduce the surface tension.
The ball is stressed as shown. In the picture: FW is the force of the micro-clip, fz is the weight of the ball, ft is the tangential adhesion of the ball to the face, fn is the normal adhesion of the ball to the face, and Fn is the normal adhesion of the ball to the pipette. .
Control Unit System Schematic Vacuum microclips rely on the pressure difference Ap between the system itself and the atmosphere to pick up and release the ball. Therefore, before the ball is picked up and released, the pressure difference must be determined. The aerodynamic force Fw=AApA applied to the vacuum microclip mechanical analysis chart is the area of ​​the pipette.
The condition for picking up the ball is that the condition for releasing the ball is to determine the magnitude of the pressure difference required to pick up and release the ball, respectively, by equations (1) and (2).
Experiments showed that the suction pipe tip length to take 3 ~ 5mm, suction pipe diameter to take 1mm, suction angle between the pipe and the working platform to take 45 * can be higher suction, placement success rate of 121. 3 fuzzy controller ball release process, The basic control strategy of the force exerted by the vacuum micro-clip is to gradually increase the working pressure of the vacuum micro-clip and to accurately stabilize the pressure value in the system-calibrated pressure value, without over-regulation or allowing only a slight overshoot to avoid “blowingâ€. Run the ball.
For an electrically proportional pressure valve with high linearity, the output pressure Pp is considered to be proportional to the input current i, ie Pp =ki,ki is a proportionality constant. Thus, the operating pressure of the vacuum microclip is adjusted by changing the magnitude of the proportional valve input current i.
Among them, Pu is a pressure loss. Ignore the micro-clip pressure loss caused by pipe bending and other factors, the micro-clip pressure loss can be approximately equal to Ploss = Pv2W (2Di), P: air density; the air flow rate in the pipe; the resistance loss coefficient along the line; the pipe length, The diameter of the pipe.
There are many factors affecting the coefficient of resistance loss X along the way, which is difficult to determine. At the same time, due to the strong nonlinearity of the pneumatic system and the gas compressibility, traditional controllers based on precise mathematical models are difficult to control the pressure of the vacuum microclips. adjust. This article uses a fuzzy logic controller to control the pressure. It is a block diagram of the vacuum micro clamp pressure control system based on fuzzy logic control.
The two inputs to the controller are the pressure deviation e() and the derivative gain, e(() = p, ui(t)*pi, where pi is the standard work calculated by the PC based on the ball size and the robot's operating environment. Pressure: The input of electrical proportional pressure valve u (t) = x ((), ux (() is the output after the fuzzy controller is unambiguous, and ku is the output gain. The controller is specifically designed as follows.
3.1 Fuzzy () language variables. The linguistic values ​​of EEC and U are taken as NB, NS, Z, PS, PB.NB, NS, Z, PS and PB respectively mean: negative large, negative small, zero, positive small and positive.
The triangle function is chosen as the membership function of the language value. The form of the triangle function is as follows: 3.2 The fuzzy logic rule base The design idea of ​​the fuzzy control rule is: When the deviation is large, the change of the control quantity should try its best to make the deviation rapidly decrease; when the deviation When smaller, the deviation is eliminated on the premise of stable system. Avoiding and suppressing overshoot are the principles that must be emphasized in the design. The definition of the fuzzy logic rule is as follows: EisAandECisBthenUisC. Meaning of EEC and U in the sentence As stated above, A, B and C are the language values ​​of the linguistic variable. According to the language value of each language variable, the controller rule base has a total of 25 rules. The specific rules are shown in Table 1. Table 1 Fuzzy logic rule library table The first column is the language variable value of E, the first is the EC language variable value, and the rest is the U language variable value.
3.3 Fuzzy decision-solving methods Common methods used in touch-up are: center of gravity, maximum degree of membership, and median judgment. Select the famous Mamdani reasoning method MIN-MAX-center of gravity method. The specific content of this method is in the deviation change rate! Neutralization A did not perform any experiments for entry into the lab. xnki.net validated the fuzzy controller's regulation effects, designed the fuzzy system using MATLAB's FuzzyLogic toolbox, and made the first step in the simulation, first determining the languages. The range of variables.
Set Ku = KP = Kd = l, proportional valve outlet pressure change range of 0.05 ~ 0.9MPa determine the pressure deviation e () change range of a 0.9MPa ~ 0.9MPa; limit deviation rate of change in the range of a l ~ 1; Fuzzy controller output range of a l ~ 1. Simulation results show that the use of fuzzy logic controller has a good control effect.
China Education Research Grid uses the rich resources of China Education and Research Online to build the largest, most advanced and most practical grid in China and even in the world. Because China University’s “ChinaGrid*†is currently under construction led by our university, Prof. Jinhai said that every time he talks about the grid, he is extremely proud. The students heard such exciting news as being in Central China. The pride of the HKUST also arises spontaneously.
Polyethylene Pellets,Polyethylene Raw Materials,Polyethylene Acid Resistance,Adhesive For Polyethylene
Henan Kepeiao New Materials Co.,Ltd. , https://www.kopeochem.com