Servovalve Hydraulic Motor Speed Control System Performance Studies

With the development of control theory and computer technique, valve-controlled motor velocity servo control system has been widely used in many areas, such as: mine, port, factory, building site, ocean and so on. The system's speed can't keep stable when its load changes. And some parameters may change during working process, so valve-controlled motor system is a typical complex nonlinear system. Therefore, at present, it is an important task for controlling system's speed and conquering the interferences caused by the changes of load and its model.In the paper, firstly, the math model of valve-controlled motor module of experimental facility was established and analyzed. Based on this, the paper introduced the basic principle of PID controller and fuzzy controller, adopted integral apart PID and fuzzy algorithms to control the experimental system, then analyzed and compared the simulation results. At last, a new control scheme based on fuzzy Gaussian neural network(FGNC) for valve-controlled motor was put forward. By using the control strategies of fuzzy logic and neural networks, a new control algorithm was designed to solve the control problems. The FGNC controller combined fuzzy control and neural network technique. It possessed not only the simplicity and the nonlinear control ability of fuzzy control but also the learning and adaptive functions by using neural network. Neural network is used to achieve fuzzy control reasoning and to create, correct and optimize fuzzy rules between input and output by learning from training data. FGNC controller can raise the adaptability and performances of valve-controlled motor system.Simulated and experimental results in MATLAB showed that the control system based on FGNC can achieve better control effects than that on PID control or fuzzy control in terms of adaptability and steady-state accuracy. The system hold stability and satisfied application requests. The success in debugging the FGNC can provide favorable reference in interrelated research and experiment.