Analysis Of The Pressing Characteristic And Design Of Control System For Punching Press Machine In CNG System

With the increasing price of oil and the exhaust pollution of vehicle on theenvironment, governments have paid close attention to the development of naturalgas. Compressed natural gas cylinder (CNG) is one of the key components todevelop the natural gas vehicles. However the quality requirements for CNG and therequired precision are increasing, since natural gas is high-pressure、explosive andflammable. The control accuracy of seamless steel cylinders, with high cost and lowefficiency, using conventional heating to close at the end and shut manufacturingprocess, depends on the accuracy of seamless steel tubes. In this paper, the CNGcylinder in the vehicle with punching—stretching method of production, using barheating—punching—drawing etc., which changes the plasticity of metals, makingdense metal, grain refinement, improves mechanical properties. Punch pressingmachine is an important machine in the drawing production line. Punching quicklyand accurately affects thickness of the bottom and wall of the cylinder, limiting thepressure resistance of the cylinder. A deep research on the punch pressing machinewhose key control part can be reduced into the electro-hydraulic position servocontrol system is done and the result as follow:Firstly, the history and research status of CNG producing system、pressingmachine、 electro hydraulic servo system, the application and the control inelectro-hydraulic servo system are summarized based on a number of relevantliteratures in this paper. The structure of the pressing machine and electro-hydraulicservo system is paid close attention and the mathematic model is designed.Secondly, a new optimal iterative learning control method, with which the errortrajectory can be pre-specified, in the sake of the optimal error and energy for thenonlinear systems is introduced, since the system the article is supposed to study isuncertain, repeat and with unknown disturbance etc. Proposed method in arbitraryinitial conditions can be expected that the actual error can coincides withpre-specified error trajectory over the entire interval. Inverse optimal control can optimal the system in the cost function. Adaptive ILC is for the uncertainties.Combine the adaptive iterative learning algorithm and the optimal control algorithmto realize the system optimal. The proposed tracking error of iterative learningcontrol scheme allows the system to start at arbitrary point; the method withdifferent situations and solutions is discussed systematically in the paper. The givenerror curve doesn’t have to be the monotone decreasing; as long as the situation ismet which the initial value of secondary error is to be zero.Thirdly， the problem of stabilization for a class of nonlinear with unknownparameters in state vector is studied. Make a research in a given error withinterference during iterative learning, and it can completely follow the pre-specifiederror; the system with the constant and time-varying parametric uncertainty is copedin by designing a hybrid learning scheme. The uncertain of system is separated intoconstant part and time-varying part. The method works effectively to follow thetarget trajectory with disturbance. As for the point that can not be learned throughthe traditional method, a new method is put forward to solve the problem. It is shownthat with the adoption of hybrid learning, the boundedness of all the signals inclosed-loop system is guaranteed and the tracking error coincides with thepre-specified error trajectory over the entire interval. The effectiveness of theproposed scheme is verified with theoretical and numerical results presented.Finally, the CNG system is introduced. The network structure and the hardwaresystem of control system for the CNG system are established, and the softwareprogram of principal computer and subordinate computer are designed.