CNC servo system is an important part of CNC machine tools, used to realize the feed servo control of CNC machine tools and spindle servo control.The function of the numerical control servo system is to receive the command information from the numerical control device, after power amplification, plastic processing, into the machine tool executive parts of the linear displacement or angular displacement movement.Because the CNC servo system is the last link of CNC machine tools, its performance will directly affect the accuracy and speed of CNC machine tools and other technical indicators.Therefore, the servo driving device of cnc machine tool requires good quick response performance, accurate and sensitive tracking of the digital command signal sent by the nc device, and can faithfully execute the command from the nc device, so as to improve the dynamic following characteristics and static tracking accuracy of the system.The measuring element detects the actual displacement value of each coordinate axis of the cnc machine tool and inputs it into the nc device of the machine tool through the feedback system.

Numerical control servo system is an automatic control system with mechanical displacement as the direct control target. It can also be called position follow-up system, or servo system for short.There are two main kinds of CNC machine tool servo system: one is the feed servo system, it controls the machine tool each coordinate axis of the cutting feed movement, mainly linear movement;The other is the spindle servo system, which controls the spindle cutting movement, mainly rotating movement.The control methods of servo system are mainly divided into three kinds: open loop, closed loop and semi-closed loop.It actually refers to the servo system to achieve position servo control in three ways.

Servo system development process:

Due to the characteristics of the permanent magnet synchronous motor (PMSM), which is coupled, time-varying and nonlinear, it is difficult to control the PMSM and obtain better speed control performance.Until 1971, the vector control theory proposed by Dr. F.blaschke of Siemens in Germany made a qualitative leap in the control theory of ac motor for the first time.Vector control the vector transform method was adopted, by ac motor, the magnetic flux and torque decoupling, the control of the permanent magnet synchronous motor is similar to the control of dc motor, thus greatly improve the control performance, become the basic control method of ac drive, the permanent magnet synchronous motor has good performance of speed control and position control performance, thus in the cnc machine tool feed servo system has been widely used.The ac servo system of CNC machine tools widely adopts three loop (current loop, speed loop and position loop)PID control technology, which has been productized and serialized.However, there are still some problems in the traditional three-ring PID control mode of permanent magnet synchronous motor in the application of CNC machine tools:

The parameter setting of the regulator is complicated and the error is large.The traditional manual design of the servo system regulator parameters needs to simplify the system, resulting in increased error, and the system does not work under the optimal state.

The decoupling control of the servo system needs an accurate mathematical model of the system. It has a strong dependence on the system parameters (such as the torque coefficient of the motor, the structure of the mechanical system, the size and frequency of the cutting force, etc.). When the parameters change, the performance of the system may become poor.

In the traditional research results, most of the research objects only consider the control of the motor, rarely consider the matching of mechanical and electrical parameters and the influence of dynamic cutting force on the dynamic performance of the servo system during machining.In the feed servo system of fully closed-loop nc machine tools, the mechanical feed system and the physical cutting process are contained in the position loop, and they are not completely separated subsystems from the electrical servo control system, but a new integrated electromechanical system is formed by coupling the feedback loop.

In order to improve the overall performance of CNC machine tools, experts and scholars on the CNC ac servo system, a high-level, complex, comprehensive system, has carried out extensive and fruitful research, mainly including:

The research on the compensation control of the nonlinear influence factors of the servo system of nc machine tools is very fruitful, and many theoretical achievements have been applied in practice.

In order to greatly improve the performance of the servo system, a new control mode is adopted for the high precision and high performance CNC servo system, and modern control methods such as neural network, expert system, adaptive control, robust control and fuzzy control are introduced into the control of the servo system.

The performance of the servo system is improved by using composite control strategy.In fact, each control strategy has its advantages and some problems.Therefore, the interpenetration and combination of various control strategies can overcome the shortcomings of a single strategy, improve the control performance, and better meet the requirements of the servo system of CNC machine tools.There are two main forms of composite control strategy: one is to adopt a new control strategy based on the traditional PID control strategy; the other is to adopt two or more new control strategies.The research emphasis is on the combination of neural network control and fuzzy control. Compound control will be a trend in the future.

System online identification.Ac servo system for CNC machine tools, including the permanent magnet synchronous motor parameters (constant torque, stator resistance and the stator inductance, etc.) will change in the operation, or change is the mechanical part of the system properties, cutting parameters can also be changed, these situations for numerical control machine tool system is inevitable.This will lead to the poor quality of the general controller designed according to the exact parameters, resulting in the degradation of the system performance.Adaptive algorithm and neural network control are used to identify the parameters on line, and adjust the regulator parameters according to the identification results, which has achieved certain results in improving the performance of the system.The method of parameter estimation is simple in design, small in computation and fast in convergence

The basic principle of PID controller:

In the design of any control system, the indexes of stability, dynamic characteristic, steady-state characteristic and robustness should be considered.

Stability: this is the most basic requirement of control system design.The stability of control system can be divided into internal stability and external stability.The so-called internal stability of the system is that the trajectory starting from the equilibrium point at any initial state converges to the equilibrium point when time is infinite;The external stability of the system is the stability of input and output, that is, the bounded input can get the bounded output.

Dynamic characteristics: that is, the form and speed of the system operation transition process, including response speed and overshoot.The response speed of the system can be expressed by the time that the transition process of the system has gone through.The overshoot is the maximum oscillation amplitude of the system.Generally speaking, different systems have different requirements for dynamic characteristics. For the CNC servo system, the faster its response speed is, the smaller the system tracking error is, and the higher the control accuracy will be.Steady-state characteristic: that is, when the system reaches a stable state after the transition process, the steady-state value of the controlled quantity is consistent with the expected value.For any practical engineering system, there is always an error between the steady-state value of the controlled quantity and the expected value, which can be called the steady-state error, due to the influence of the system structure, external interference, internal friction and other nonlinear factors.Steady-state error is an important symbol to measure the control accuracy of the control system, and there are generally specific requirements in the technical indexes of the control system.

Robustness: when the constraints of the system change, the functional characteristics of the system will not be affected.If the robustness of the system is good, the system can still maintain its stability when the parameters change.In the process of transition, the response speed and overshoot of the system are not affected by the change of parameters.The parameter change mentioned here includes not only the actual external parameter change, but also the system internal parameter change.

PID control technology is one of the earliest control strategies developed, which has a history of several decades.It is widely used in industrial control due to its advantages of simple algorithm, good robustness, high reliability and easy adjustment.When the structure and parameters of the controlled object cannot be fully mastered, or an accurate mathematical model cannot be obtained, the structure and parameters of the system controller must be determined by experience and on-site debugging. At this time, the application of PID control technology is most convenient.In practical engineering applications, PI control and PD control can also be used as required.PID controller is based on the system deviation, through proportional, integral and differential operations to control the amount of adjustment.