Study On Control Methods Of Doule Servo Valve-Controlled Cylinders' Synchronization

Synchronous hydraulic control industry has been an important issue. In the multi-cylinder hydraulic system, there are many factors that can affect the synchronous precision, such as the external load of hydraulic cylinders, the leakage, the friction resistance, the manufacturing precision, the gas melting in the oil and the elastic deformation of the structure. With the development of aerospace technology and modern machining technology, metal machining equipment, metallurgical machinery, engineering machinery, aerospace drivers and so on call for high-precision hydraulic synchronous drive technology urgently, which also highlights the importance of the study for hydraulic synchronous control methods.In order to obtain the high-precision control of the synchronization, the hydraulic cylinders are usually controlled by the traditional servo-valves. But in a common servo-valve, generally there is a certain dead-zone and the flow gain varies with the differential pressure of the valve port, which make the valve-controlled hydraulic cylinders' Synchronization system have the characteristics of nonlinear and time-varying. This makes the traditional control strategy -- PID control difficult to reconcile the contradiction between rapidity and stability. The robustness the system is not good enough either. Thus it's difficult to obtain good synchronization control. Traditional PID controller can not tune the parameters for real time, so the parameters of the controller can not change with the variation of actual status, which leads to control deterioration of the system and poor system adaptability.In the paper, firstly, with the deep study on the valve-controlled cylinder module of Experimental Facility, the mathematical model of the system was established. Then the introduction of fuzzy control and neural network control was followed. With the combination of the PID controller and the two artificial intelligent control methods, a fuzzy self-tuning PID controller, a neural network self-tuning PID controller and a fuzzy neural network self-tuning PID controller (FNN) were established. All the controllers achieved good control performance. Particularly the FNN, combined fuzzy control technology and neural network technology, possessed not only the simplicity and the nonlinear control ability of fuzzy control but also the learning and adaptive functions by using neural network. So the adaptivity and the control performance of the valve-controlled system capabilities were improved. The simulation results showed that these three control methods which helped the system overcome the inherent time-varying, nonlinear of the system and the outside interference, achieved the self-tuning of the three parameters--ratio, integral and differential coefficient, and thus achieved good control accuracy. The success of the FNN controller debugging can provide favorable reference in interrelated research and experiment.