Nc Test Platform Based On Repetitive Control Servo Adjustment

A conventional PID technology is widely used in the NC system due to its simple structure, easy of turning methods and no relation with accurate models of controlled objects. But it is difficult to yield a good control performance because its three parameters can get not optimal values but compromised values. Furthermore, these parameters are turned in the unloaded condition, so the control performance will be affected if the load of controlled object or environment condition changes deeply.In this paper, a new adjusting technology applied for NC system is proposed. This technology based on the analysis of the repetitive control theory has a good dynamic-trace capability and strong robust. This paper focuses on the following approaches:1. We introduced all parts of a general repetitive control system after depicting the regulations of the repetitive control theory. The parts include inner model, delay section, compensator, low pass filter and controlled object. And we also analyzed the robust, convergence, following error of the repetitive control system.2. Based on choosing its hard wares, designing the cabinet and soft wares we built up the experimental NC system.3. Using the M sequence as the system's input we identified the model of the controlled object of the experimental NC system. In order to get the model, we firstly gathered the input and output of the system with the gather-data capability of PMAC, Secondly we used the Matlab software to deal with the gathered data and got the model. Thirdly, the model having unstable zero points was adjusted and got the stable adjusted model.4. We depended on the extended user servo function of PMAC to make the repetitive servo algorithm install in the experimental NC system. The tasks include the analysis of the extended servo technology of PMAC, the design of the servo system of the experimental NC system, writing the program of the user servo algorithm and the execution of the program in the experimental NC system.5. In order to check the effect of the designed repetitive control servo system we simulated it with the Matlab software and did some experiments in the experimental NC system. From the results of the simulation and experiment, we can conclude that the repetitive control servo algorithm has the good control performance and it can reduce the following error of the system to 2%.