Simulation Study of CNC Servo System Based on Fuzzy Theory

Abstract: In CNC servo systems, when system parameters change or are subjected to external disturbances, PID control can easily lead to excessive overshoot or even oscillations, thus degrading the dynamic and static performance of the system. This paper proposes a fuzzy PID controller, which combines the advantages of PID control and fuzzy control, exhibiting better adaptability and anti-interference capabilities. Taking the Z-axis position servo system of a high-speed CNC gear milling machine developed in this institute as the research object, simulations were conducted in Matlab using both PID and fuzzy PID control. The results show that the fuzzy PID controller has stronger robustness and anti-interference capabilities, and better meets the performance requirements of CNC servo systems.

Keywords: Fuzzy PID controller; AC servo system; position control; MATLAB simulation

Abstract: Variation of the system parameters and external disturbances always happen in the CNC servo system. With a PID controller, it may lead to large overshoot even oscillation, then the steady state and dynamic state performance will be worse. In this paper, a fuzzy-PID controller was proposed, and it incorporated the advantages of PID control and fuzzy control, which had good performance of adaptation and disturbance rejection. Based on the z-axis position servo system of a high-speed milling machine researched on our institute, simulations were carried out in Matlab with PID controller and fuzzy-PID controller. The results show that the system has stronger robustness and disturbance rejection with the latter, which can meet the performance requirements of the CNC servo system better.

Keywords: Fuzzy-PID controller; AC-servo system; Position control; MATLAB simulation

1. Introduction

The servo system of a CNC machine tool is a crucial component connecting the CNC system and the main body of the CNC machine tool. Its performance directly affects the stability, response speed, and tracking accuracy of the CNC machine tool. Maintaining good dynamic characteristics and steady-state tracking accuracy in the AC servo system of a CNC machine tool, even under the influence of time-varying and nonlinear AC motor parameters and load disturbances, is an important challenge in servo system research.

Traditional linear PID control is widely used in AC servo systems, but its poor parameter adaptability and weak anti-interference capability make it difficult to meet the requirements of CNC servo systems. Fuzzy control, on the other hand, is an intelligent control method that simulates human logical reasoning rules and does not rely on a precise mathematical model of the object. Furthermore, it is insensitive to changes in the parameters of the controlled object and has strong robustness, making it well-suited to overcome the influence of nonlinearity, time-varying factors, and coupling in AC servo systems. This paper addresses the drawback of low steady-state accuracy in fuzzy control by combining PID control with fuzzy control to propose a fuzzy PID controller. This controller combines the flexibility and strong anti-interference capabilities of fuzzy control with the high steady-state accuracy of PID control.

2. Structure of the servo system

Numerical control AC servo systems typically employ a three-loop structure with position, speed, and current feedback. This three-loop structure enables the system to achieve better static and dynamic performance.

The current loop and speed loop are the inner loops, and the position loop is the outer loop. The position loop's function is to ensure the system's static accuracy and dynamic tracking performance, enabling the entire servo system to operate stably and efficiently. Therefore, the position loop is crucial to the entire servo system design. This paper uses a fuzzy PID controller as the position regulator (both the current loop and speed loop use PI regulators) to meet the requirements of fast, overshoot-free position control in CNC servo systems.


For details, please click: Simulation Research of CNC Servo System Based on Fuzzy Theory