Noise is generally defined as the unwanted component in a signal, and it is ubiquitous. In CNC machine tools and their surrounding environment, noise disturbances are unavoidable. These disturbances include drift caused by temperature changes and various electrical disturbance signals. All noise disturbance signals inevitably reduce the tracking accuracy of the servo system. In CNC machine tool control cabinets, grounding, shielding, and isolation technologies are generally used to eliminate the impact of noise disturbance signals.
Designing disturbance observers for various disturbance signals and compensating for them in servo control systems is a way to reduce the impact of disturbances and thus improve system robustness. Many scholars both domestically and internationally have conducted research on compensation control methods for disturbance signals in servo control. KIM et al. designed a fuzzy disturbance observer for feedback tracking control of multi-input multi-output systems and applied it to the speed control of permanent magnet synchronous motors; RYOO et al. designed a robust disturbance observer and conducted experiments in track tracking control of optical disc drive systems; LU et al. used sliding mode repetitive control theory to study disturbance signal observers; Dong Mingxiao et al. combined hybrid sensitivity design methods to design a robust H∞ controller for CNC machine tool servos.
This paper analyzes the impact of noise disturbance on the tracking accuracy of a servo system and proposes a control method for observing and compensating for noise disturbance signals. The noise disturbance is observed by detecting the voltage applied to the servo driver and the angular displacement of the servo motor, and the disturbance compensation is superimposed on the position controller output to achieve compensation. Simulation experiments were conducted for typical sawtooth wave noise disturbance signals.
CNC servo system model and the impact of electrical disturbances
A simplified block diagram of a semi-closed-loop feed servo system with noise disturbance. Let the position command signal from the interpolator be X(s), and the angular displacement output signal of the servo motor be Y(s). Assume that the position control loop adopts proportional control, and the transfer function is given. The steady-state error caused by the noise disturbance signal is related to the disturbance signal itself, as well as the part of N(s) before the point of action in the feed servo system.
Noise disturbance observation and compensation methods
In the feed servo system, a noise disturbance observation and compensation stage is added. As shown in Figure 2, the disturbance signal N(s) is observed by detecting the voltage signal applied to the servo driver and the angular displacement of the servo motor, and the disturbance compensation amount is superimposed on the position controller output to achieve compensation.
From equations (3) to (5), we can obtain the closed-loop transfer function G(s) of the system after adding noise disturbance and observation and compensator, which is completely consistent with equation (2). This shows that the observation and compensation method for noise disturbance shown in Figure 2 can compensate for the disturbance effect and improve the system's anti-interference ability.
Simulation of Noise Disturbance Observation and Compensation Methods
In the position controller stage, PID control is adopted with a proportional coefficient of 8.1, an integral coefficient of 0.002, and a derivative coefficient of 0.032. When conducting observational compensation simulation studies on noise disturbances, the position command input signal is assumed to be 2sin(0.4πt); the noise disturbance is a sawtooth wave signal with an amplitude of 0.5 and a period of 2s.
When noise disturbance is not considered, the tracking error of the servo feed system is shown in Figure 4, which is within ±0.006 mm. When noise disturbance is added but no disturbance observation and compensation are performed, the tracking error is shown in Figure 5, which is within ±0.02 mm. When the noise disturbance observation and compensation method described in this paper is used, the tracking error is shown in Figure 6, which is within ±0.007 mm. The comparison demonstrates that the noise disturbance observation and compensation method studied can effectively improve the anti-interference capability of the servo feed system.
in conclusion
Noise signals are ubiquitous. At the interface of the servo system driver in CNC machine tools, noise disturbances include drift caused by temperature changes and various electrical disturbance signals. These noise disturbances inevitably reduce the tracking accuracy of the servo system. This paper proposes a software compensation method for noise disturbances, rather than focusing on hardware: by detecting the voltage applied to the servo driver and the angular displacement of the servo motor, the noise disturbances are observed, and the compensation amount is superimposed on the position controller output. Simulation results for typical sawtooth wave disturbance signals show that the proposed observation and compensation method can effectively improve tracking accuracy and enhance the system's anti-interference capability. This method is a valuable supplement to hardware-based anti-disturbance techniques.
