CN100520650C - Fuzzy PID control method and execution apparatus of numerical control machine - Google Patents

Abstract

The invention discloses a fuzzy PID control method and a realization device of a numerically controlled machine tool. It adopts a man-machine interface, task coordination, motion control and a programmable controller, and loads the motion control and the programmable controller into the kernel space of the system, and the man-machine interface , The task coordination part is loaded in the user space of the system, and communicates with each other through the shared communication buffer of the system; the motion control includes two parts: motion planning and axis control; among them: the axis control adopts the fuzzy PID control method according to the setting value of the motion planning Dynamically determine the control parameters of the axis and realize the stable control of the axis, thus forming the movement of the servo axis of the machine tool during the machining process. By adopting the invention, the method of combining fuzzy control and PID control can be used to support a practical intelligent numerical control device with excellent performance, so as to achieve the optimal control of the motion of the servo axis of the machine tool in the process of processing.

Description

Fuzzy PID Control Method and Realization Device for Numerical Control Machine Tool

technical field

The invention relates to the control technology of a numerical control device, in particular to a fuzzy PID control method and a realization device of a numerical control machine tool, which can realize intelligent self-adaptive adjustment of control parameters of a machine tool servo axis.

Background technique

In recent years, with the development of manufacturing technology, the CNC machining process and processing parts have become increasingly complex. The control requirements of complex parts and processes make it difficult for existing device control methods to meet system performance requirements. Take the PID (P for proportional, I for integral, D for differential) control method adopted by the existing numerical control device as an example to illustrate. The PID control method has a long history and is widely used in the axis control of numerical control devices because of its simple principle and easy implementation. The PID controller using the PID control method has three important engineering parameters K _p (proportional parameter), K _i (integral parameter) and _{K d} (differential parameter). Once these parameters are set, they will not change. The controller uses the same parameters to deal with different execution situations, and does not adjust with the change of the controlled parameters. Therefore, it is suitable for deterministic control systems that can establish accurate mathematical models. However, due to the complexity of the machining process, the servo axis of the machine tool is a time-varying dynamic system. Machine tool servo axes are often switched between different motion states, such as rapid movement, inching, roughing, and finishing. Different motion states have different performance (speed, precision) indicators and require different control parameters. Due to the difficulty in setting the parameters of the PID controller using the PID control method, the ideal control effect cannot be achieved for the CNC machine tool.

Aiming at the problems of PID controller parameter tuning difficulties, poor performance, and poor adaptability to operating conditions, many control methods have been developed in recent years, such as system identification, adaptive control, and robust control. However, none of the above three methods can get rid of the quantitative idea based on the mathematical model of the controlled object in essence, and it is difficult to effectively and accurately control the complex nonlinear uncertain system. The emergence of intelligent control represented by fuzzy control, neural network and expert control, especially the rapid development of fuzzy control technology in recent years, provides new ideas for solving such problems. In intelligent control, the object studied by the system is no longer the controlled object, but the controller itself. The controller is no longer a single mathematical analysis model, but a generalized model combining mathematical analysis and knowledge system. Therefore, the application of intelligent control technology to develop intelligent numerical control devices with anthropomorphic intelligence features to achieve optimal control of process parameters such as feed speed, cutting depth, coordinate movement, and spindle speed has become the development trend of numerical control devices.

The development of intelligent control technology has gone through a period of time, but it is still in the pioneering research stage, and there are many theoretical problems that have not been resolved. For example, the commonly used two-dimensional fuzzy controller has good dynamic performance, but poor static performance. Steady-state errors cannot be eliminated. These problems make intelligent control technology far from being popularized in industrial process control as it is in household appliances. The engineering practice of intelligent control is faced with how to solve the problems of "controller design with better control performance than traditional PID" and how to facilitate the "simplicity" of system implementation.

Contents of the invention

The object of the present invention is to provide a fuzzy PID control method and an intelligent realization device for numerically controlled machine tools. The fuzzy PID method used in the present invention is a design method that supports practical and high-performance intelligent numerical control devices by combining fuzzy control and PID control methods; based on this method, the present invention uses a PC platform to develop an intelligent numerical control device. Realize the device to realize the control optimization of the servo axis movement of the machine tool during the machining process.

In order to achieve the above object, the technical scheme of the present invention is as follows:

Using human-machine interface, task coordination, motion control and programmable controller, the motion control and programmable controller are loaded in the kernel space of the system, and the human-machine interface and task coordination are loaded in the user space of the system, and they are connected to each other through the system The shared communication buffer communicates with each other; the motion control includes two parts: motion planning and axis control; the axis control adopts the fuzzy PID control method to dynamically determine the control parameters of the axis and realize the stable control of the axis according to the set value of the axis motion planning, thus forming a CNC machine tool Movement of the servo axis during machining;

The fuzzy PID control method is composed of two parts, a parameter-adjustable PID regulator and a fuzzy self-tuning mechanism, wherein: the PID regulator is used as the basic control unit, aiming at the nonlinear characteristics of the machine tool servo axis during processing and various processing Requirements, the introduction of fuzzy control technology to achieve the adjustment of regulator parameters. The fuzzy self-tuning mechanism receives the setting value of the motion planning output and detects the actual position value of the shaft through the encoder to form the error e and the error change rate ec, which is then processed by the fuzzification in the fuzzy self-tuning mechanism to form a language variable. Inference and defuzzification processing generate the parameters of the PID regulator, under the control of the parameters, the PID regulator controls the motion of the servo shaft of the machine tool, thereby completing the control process of the servo shaft; the fuzzy reasoning is based on the error e and the error change The rate ec is used as input, and the inference rule has the following form:

If(e is...)and(ec is...)then(K _p is...)(K _i is...)(K _d is...) (1)

Based on the rules, it is concluded that the PID parameters of the PID regulator are nonlinear functions of the error e and the error change ec, which can be specifically recorded as:

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; p</span>}}$

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; d</span>}}$

分别表示PID参数中的比例、积分、微分的初始值，{e_j+ec_j}_p、{e_j+ec_j}_i、{e_j+ec_j}_d表示经模糊推理和解模糊化后得到的调整PID参数中比例、积分、微分的变化量；其中j为序数；in,

Respectively represent the initial values of the proportion, integral and differential in the PID parameters, {e _j +ec _j } _p , {e _j +ec _j } _i , {e _j +ec _j } _d represent the obtained after fuzzy reasoning and defuzzification Adjustment of the proportional, integral and differential variation in the PID parameters; where j is an ordinal number;

The implementation device of the fuzzy PID control method of the numerically controlled machine tool is based on the ISA bus and powered by a power supply, including: a display, a central controller, a machine tool operation panel, an axis control circuit and an interface circuit, wherein: the central controller and the display, the machine tool operation panel Communication, and communicate with the axis control circuit and interface circuit through the ISA bus, and the control program is installed in the FLASH or DOM of the central controller to realize the fuzzy PID control of the CNC machine tool;

The control program includes axis control (Axisctrl), axis characteristics (AxisProperty), axis constraint (AxisConstraint), axis setting (AxisSetpoint), axis detection (AxisSense), axis drive (AxisAction), and axis fuzzy PID control (AxisFuzzyPID) parts; The axis control is the main program, in which: the axis characteristics record the dynamic characteristics and static characteristics of the axis; the axis constraints record the movement ability of the axis; the axis setting is used as the interface between fuzzy PID control and motion planning, and the axis setting position is recorded; the axis detection is used as the position code The interface of the controller detects the actual running position and speed of the axis; the axis drive serves as the interface of the servo axis to provide the signal of the motor running; the axis fuzzy PID control uses the fuzzy PID control method to call the axis characteristics, axis constraints, axis detection, and axis drive. Data, realize the optimal control of axis motion;

The fuzzy PID control includes: initialization operation, input data operation, error judgment operation, fuzzification operation, fuzzy inference operation, defuzzification operation, gain control operation, control protection operation, PID control operation, fuzzy PID control operation, wherein:

Initialization operation (fuzzyInit()), specifically including PID control operation ratio, integral, differential parameter initialization and fuzzy self-tuning parameter initialization;

Input data operation (inputError()) reads the setting value (setPoint) given by the motion planning from the axis setting (AxisSetpoint), reads the actual position value of the current axis from the axis detection AxisSense, and forms the error e and error change rate of the fuzzy PID control ec;

The error judgment operation (ctrlAccuracy()) evaluates the execution of the current PID control operation; if the value of the input error e is less than the control accuracy of the system by an order of magnitude, no adjustment is required. At this time, skip the fuzzy control related modules and directly run the PID control Operation; otherwise, run fuzzy control related modules and adjust related parameters of PID control;

The fuzzy operation (fuzzy()) maps the input value from the input data operation into a fuzzy value in a certain interval according to the fuzzy method and the fuzzy membership function. The domain of fuzzy value is:

e={-0.75, -0.5, -0.25, 0, 0.25, 0.5, 0.75}

ec={-0.15, -0.1, -0.05, 0, 0.05, 0.1, 0.15}

The corresponding fuzzy subset is:

e, ec={LN (the negative direction is too large), MN (the negative direction is centered), SN (the negative direction is small), ZZ (zero), SP (the positive direction is small), MP (the positive direction is centered), LP ( The positive direction is too large)}

The fuzzy inference operation (fuzzyInf()) derives the parameters K _p , K _i , K _d corresponding to the input error e and the error change ec according to the fuzzy inference rules and the fuzzy inference method, and the output value is in the form of a fuzzy value;

The defuzzification operation (deFuzzy()) defuzzifies the output value of the fuzzy inference operation according to the output membership function, and obtains the proportional, integral, and differential adjustment gain parameters (Δp, Δi, Δd) of the PID control operation;

The gain control operation (gainCtrl()) adjusts the proportional, integral, and differential parameters K _p , K i , and K d of the PID control operation according to the value of the proportional, integral, and differential adjustment gain parameters Δp, _Δi , and _Δd obtained by the defuzzification operation value, the adjustment principle is:

{e _j +ec _j } _p ＝p _a ×Δp

{e _j +ec _j }i=i _a ×Δi (3)

{e _j +ec _j } _d =d _a ×Δd

Among them, p _a , i _a , d _a are real numbers greater than zero, and j is an ordinal number;

The control and protection operation (ctrlSafe()) sets the upper limit of the adjustment of the proportional, integral and differential parameters of the PID control operation, and the variation range of the proportional, integral and differential parameters of the PID control operation: if the proportional parameter K _p of the PID control operation is greater than the set If the value of the proportional parameter K _p of _{the control operation is equal to the set maximum value p max ;} if the integral parameter K _i of the PID control operation is greater than the set maximum value i _max , then the PID control operation The value of the integral parameter K _i of the PID control operation is equal to the set maximum value i _max ; if the differential parameter K _d of the PID control operation is greater than the set maximum value d _max , then the value of the differential parameter K _d of the PID control operation is equal to the set value The maximum value d _max ; among them, p _max , i _max , and d _max are the upper limits of the proportional, integral, and differential parameter adjustments of the PID control operation, and their specific values are set according to the actual performance indicators of the shaft, tool, and motor;

PID control operation (PIDCtrl()) realizes the control function of the axis by using the traditional PID control method according to the values of the proportional, integral and differential parameters K _p , K _i , and K _d obtained from the control and protection operation;

The fuzzy PID control operation is to call the above input data, error judgment, fuzzification, fuzzy reasoning, defuzzification, gain control, control protection and PID control operation to realize the optimal control of the servo axis of the machine tool; the specific process is: firstly, input data Operation, after the input data operation, the error judgment operation is performed. If there is an error, the fuzzification operation, fuzzy reasoning operation, defuzzification operation, gain control operation are performed in sequence, and then enter the control protection operation; if there is no error, go directly to the control protection operation; Finally, perform PID control operation;

The initialization of the proportional, integral and differential parameters of the PID control operation includes the initialization of the initial value of the proportional, integral and differential parameters of the PID control operation and the initialization of the maximum value of the proportional, integral and differential parameters of the PID control operation; the initialization of the fuzzy self-tuning parameters includes the fuzzy self-tuning parameter initialization Control the initialization of input and output membership functions, the initialization of fuzzy control input and output coefficients, and the initialization of fuzzy inference rules; define the membership function of the domain error e and error change ec of fuzzy values as an open interval, and the membership function of the interval boundary Select the trapezoidal membership function, and the others are triangular membership functions; the induction of the proportional, integral, and differential parameters K _p , K _i , and K _d derivation rules: based on a large number of engineering experiments, the fuzzy reasoning method selects the product reasoning machine, and the product The inference engine uses the independent inference principle of fuzzy combination, adopts the meaning of Mamdani product, all t-norm operators select the algebraic product operator, and all s-norm operators select the maximum operator; the defuzzification operation selects the center The average defuzzifier serves as the defuzzifier.

Compared with the prior art, the present invention has the following advantages:

One is high control precision. The method makes full use of the motion characteristics of the servo axis of the machine tool in the processing process, that is, in a certain processing process, the control object is a linear system, and a PID regulator (in order to distinguish the PID controller of the prior art, here is called a PID regulator) is used as the Basic control unit. Therefore, the characteristics of simple PID method, no static error in steady state and high control precision are fully utilized, and the static error problem existing in common two-dimensional fuzzy control is avoided.

The second is good dynamic performance. Fuzzy control is introduced to switch between different motion states of the servo axis of the machine tool during the machining process. Not only can the control parameters of the PID regulator be adjusted online according to the actual operating conditions on site, so that it tends to the optimal state, but also the knowledge related to the dynamic characteristics and control strategies of the controlled object can be embedded in the numerical control implementation device to adapt to The change of the nonlinear disturbance in the axis control makes the system more robust.

The third is high execution efficiency. According to the actual operation situation on site, adjust the value of the PID regulator parameters at any time to ensure that the current PID control parameters are relatively optimized, so a higher execution speed can be obtained.

Four is high reliability. In view of the special status of the detection link and the drive link in the numerical control device, the system adopts the field programmable gate array technology to design and realize the detection and drive circuit, and improves the reliability of the system under the premise of ensuring the system is open.

Description of drawings

Fig. 1 is the overall structural diagram of the method of the present invention.

Fig. 2 is a schematic diagram of the axis control of the present invention in Fig. 1 .

Fig. 3 is a structural diagram of a device for realizing the method of the present invention.

FIG. 4 is a structural block diagram of the axis control circuit in FIG. 3 .

FIG. 5 is a functional block diagram of the field programmable gate array in FIG. 4 .

Fig. 6 is a structural diagram of the control program for realizing the servo axis control of the machine tool according to the present invention.

Fig. 7 is a flow chart of the control program for realizing the servo axis control of the machine tool according to the present invention.

Fig. 8 is a fuzzy PID control operation diagram in the control program of the present invention.

Fig. 9 is the change trajectory of the servo axis error in the fuzzy PID control of the present invention.

Fig. 10 is the change trajectory of the servo axis error under PID control in the prior art.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings.

(1) Fuzzy PID control method for CNC machine tools

Referring to Fig. 1, the human-machine interface, task coordination, motion control and programmable controller are used, and the motion control and programmable controller are loaded in the kernel space of the system, and the human-machine interface and task coordination are loaded in the user space of the system. Communicate with each other through the system's shared communication buffer; where:

The man-machine interface is used to accept operation commands and workpiece programs, to display the processing process of the realization device of the present invention, and to realize the interaction between machine tool operators and the realization device of the present invention.

Task coordination is composed of state management and part program interpretation. State management adopts finite state machine technology to realize automatic management and switching between processing states, such as automatic processing and manual processing. Part program interpretation adopts compiling technology to realize the decomposition of operation command and part program to machine tool control command.

Motion control includes motion planning and axis control. Motion planning refines machine tool control commands through speed planning and curve interpolation, and generates machine tool servo axis setting values with position and speed constraints; axis control adopts fuzzy PID control method based on setting values Dynamically determine the control parameters of the axis and realize the stable control of the axis, thus forming the movement of the servo axis in the process of CNC machine tool processing.

The programmable controller consists of two parts: the ladder diagram editor and the programmable controller engine. The ladder diagram editor implements the editing and conversion of the machine tool electrical control program represented by the ladder diagram. The programmable controller engine uses interpretation technology to realize the interpretation and execution of the control program of machine tool electrical appliances, so as to realize the control of machine tool electrical appliances, such as coolant switch and tool change.

The user inputs the workpiece program through the human-machine interface, and the workpiece program forms a machine tool control command under the control of task coordination. After the machine tool control command generates the axis control setting value through motion planning, it enters the axis control stage; the axis control stage adopts fuzzy PID control method, Adjust the control parameters online according to the motion state of the axis, so as to realize the optimal control of the servo axis.

The fuzzy PID control method is composed of two parts: a PID regulator with adjustable parameters (in order to distinguish the PID controller of the prior art, it is called a PID regulator here) and a fuzzy self-tuning mechanism. The regulator is used as the basic control unit to take advantage of its high control accuracy. Aiming at the non-linear characteristics and various processing requirements of the servo axis of the machine tool in the processing process, the fuzzy control technology is introduced to realize the adjustment of the regulator parameters. The fuzzy self-tuning mechanism receives the setting value of the motion planning output and detects the actual position value of the shaft through the encoder to form the error e and the error change rate ec, which is then processed by the fuzzification in the fuzzy self-tuning mechanism to form a language variable. Reasoning and defuzzification processing to generate the parameters of the PID regulator. Under the control of this parameter, the PID regulator controls the movement of the servo axis of the machine tool, thus completing the control process of the servo axis.

The fuzzy reasoning takes the error e and the error rate of change ec as input, and the reasoning rules are as follows:

If(e is...)and(ec is...)then(K _p is...)(K _i is...)(K _d is...) (1)

Based on the rules, it is concluded that the PID parameters of the PID regulator are nonlinear functions of the error e and the error change ec, which can be specifically recorded as:

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; p</span>}}$

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; d</span>}}$

分别表示PID参数中比例、积分、微分的初始值，{e_j+ec_j}_p、{e_j+ec_j}_i、{e_j+ec_j}_d表示经模糊推理和解模糊化后得到的PID参数中比例、积分、微分的变化量；其中j为序数。in,

respectively represent the initial values of proportion, integral and differential in the PID parameters, {e _j +ec _j } _p , {e _j +ec _j } _i , {e _j +ec _j } _d represent the values obtained after fuzzy reasoning and defuzzification The variation of proportional, integral and differential in PID parameters; where j is an ordinal number.

Although both e and ec are functions of time, the value of the PID regulator parameters is only determined by the specific values of e and ec, and has nothing to do with what time it is. Therefore, the fuzzy PID control is essentially a static (or steady) nonlinear system, which can realize the independent tuning of each nonlinear parameter. In order to ensure the stability of the CNC machine tool and eliminate the vibration that may occur during the machining process, this method limits the range of the control parameters of the PID regulator.

(2) Implementation device of fuzzy PID control method for CNC machine tools

As shown in Figure 3, the implementation device of the fuzzy PID control method for CNC machine tools is based on the ISA bus and powered by a power supply 4, including: TFT display 1, central controller 2, machine tool operation panel 3, axis control circuit 5 and interface circuit 6 , wherein: the central controller 2 communicates with the display 1 and the machine operation panel 3, and communicates with the axis control circuit 5 and the interface circuit 6 through the ISA bus, and the control program is installed in the FLASH or DOM memory of the central controller 2 to realize the control Fuzzy PID control of CNC machine tools; in order to ensure the control performance and openness of the control device, the software platform of the device adopts RTLinux, a general-purpose operating system with real-time expansion.

details as follows:

The size of the TFT display 1 is 10.4", adopts a display mode of 640 X 480, and supports 16-bit enhanced color.

Central controller 2 adopts standard industrial control board, CPU adopts Intel Pentium MMX, main frequency is 200MHz, memory is 32M, supports FLASH and DOM two ways, storage capacity is 64M.

The machine tool operation panel 3 is an interface for human-computer interaction, and completes functions such as keyboard editing, displaying information and graphics, and completing the operation of the machine tool. The operation panel comprises a main keyboard board, a function keyboard board, a key switch and an LCD indicator light, a band switch, an emergency stop switch, a cycle start, and a cycle stop button (for prior art).

The power supply 4 provides three DC power supplies of +5V, +12V and -12V for the numerical control system. It features high efficiency, high reliability, low output ripple and noise. The power supply unit is mainly composed of industrial switching power supply T-50B, power supply control board, power input board, and power supply EMI filter (commercially available products).

The core of axis control circuit 5 is Field Programmable Gate Array FPGA, and its circuit structure block diagram is shown in Fig. 4 . According to the control requirements of the servo axis of the machine tool, it provides D/A output interface, encoder input interface, switch signal interface, probe signal interface, display light signal interface, input and output circuit interface, encoder fault detection interface, ISA bus interface, And the control logic is provided by the field programmable gate array. Among them, the D/A output interface provides the control signal of the servo drive, the encoder input interface detects the position signal of the encoder in the servo motor, the switch signal interface detects the zero return point and limit position of the servo axis of the machine tool, and the probe signal interface detects the on-line of the probe The position signal and display lamp signal interface provide the execution state in the axis control, the input and output circuit interface provides the control signal for the interface circuit, and the encoder fault detection interface detects the fault of the encoder. Wherein the above-mentioned interface circuits are all standard circuits (commercially available products), and the signals generated by the circuit finally communicate with the central controller in the device through the ISA bus interface circuit, and the control software on it completes the control function of the machine tool servo axis.

The control logic of the axis control circuit is provided by a field programmable gate array, and the field programmable gate array adopts general-purpose devices. The control logic is written by the ultra-high-speed integrated circuit hardware description language VHDL according to the control function of the axis, and its functional block diagram is shown in Figure 5. The address decoding logic, bus controller, and interrupt controller provide address signals and control command signals for probe interface control logic, encoder interface control logic, encoder fault detection interface control logic, and D/A output control logic. Probe interface control logic, encoder interface control logic, encoder fault detection interface control logic, control the corresponding interface circuit according to the address signal and control command to complete the position detection function of the device. The D/A output interface control logic provides the data and control signals required by the D/A conversion chip in the interface circuit, and generates the analog voltage signal required by the servo drive. The input and output control logic and the enabling control logic provide input and output switch control signals through their interface circuits. EEPROM interface and hardware dog provide software and hardware encryption functions for the system. The control logic of the light and switch interface provides instructions and operation signals for the machine tool operating parts through its interface circuit.

The interface circuit 6 is a standard circuit (commercially available product), and its main function is to provide isolated 64 input/48 output digital interfaces, and to transmit information between the device and the functional parts of the machine tool (such as: various switches, indicator lights and relays) effect. The status of various switches (such as limit switches, etc.) in the CNC machine tool should be read into the device through the digital input port, and the indicator lights on the control panel of the CNC machine tool and the on-off of the relay are read through the digital output port of the interface circuit 6. Take control. Each channel of the interface circuit 6 is photoelectrically isolated, which enhances the anti-interference ability of the system.

Among them, the interface circuit 6 uses two D-type 37-pin sockets as input interfaces, and each socket has 32 channels; two D-type 25-hole sockets are used as output interfaces, and each socket has 24 channels. The D-type socket is connected to the machine tool electrical appliances through the terminal board, which constitutes the machine tool electrical control circuit of the device.

As shown in Figure 6-7, the control program includes axis control (Axisctrl), axis properties (AxisProperty), axis constraints (AxisConstraint), axis settings (AxisSetpoint), axis detection (AxisSense), axis drive (AxisAction), axis fuzzy PID Control (AxisFuzzyPID) part; the axis control is the main program, in which: axis characteristics record the dynamic characteristics and static characteristics of the axis, such as time constant, open-loop gain, closed-loop gain, dead zone, backlash, etc.; axis constraints record the axis Motion capability, such as maximum stroke, soft limit, hard limit, etc., is based on the safety of axis movement; axis setting is used as the interface between fuzzy PID control and motion planning, and records the axis control setting value; axis detection is used as the interface of the encoder to detect the axis The actual running position and speed of the axis; the shaft drive serves as the interface of the servo drive to provide the signal of the servo shaft motor running; the axis fuzzy PID control uses the data of the axis characteristics, axis constraints, axis detection, and axis drive according to the fuzzy PID control method to realize the axis Optimal control of movement.

Axis fuzzy PID control has the following data structure:

-PIDParam, save the value, initial value and maximum value of the three control parameters K _p , K _i , K _d of the PID regulator's proportional, integral and differential;

-fuzzyRuleSet, store fuzzy control rules in the form of tables;

-fuzzyInSet defines the input fuzzy set with a fuzzy structure;

-fuzzyOutSet defines the output fuzzy set with a fuzzy structure;

Its fuzzy structure is as follows:

struct FUZZY_STRUCT

{

double center[NUM_OF_MFS]; /*The center of each fuzzy set in the input and output universe*/

double deg_of_mbrship[NUM_OF_MFS]; /*Membership function*/

double spread; /*half the length of the fuzzy set interval*/

double gain; /* coefficient of fuzzy input and output */

double error; /* input or output value of fuzzy controller */

}

Shaft fuzzy PID control includes: initialization operation, input data operation, error judgment operation, fuzzy operation, fuzzy reasoning operation, defuzzification operation, gain control operation, control protection operation, PID control operation, fuzzy PID control operation, wherein:

The initialization operation (fuzzyInit()) includes the initialization of the PID control operation proportional, integral and differential parameters and the fuzzy self-tuning mechanism parameter initialization, specifically the initialization of the data structures PIDParam, fuzzyRuleSet, fuzzyInSet and fuzzyOutSet.

The initialization of the proportional, integral and differential parameters of the PID control operation includes the initialization of the initial value of the proportional, integral and differential parameters of the PID control operation and the initialization of the maximum value of the proportional, integral and differential parameters of the PID control operation; the parameter initialization of the fuzzy self-tuning mechanism includes Initialization of membership functions, input coefficients and output coefficients, input values and output values in fuzzyInSet and fuzzyOutSet, and initialization of fuzzy inference rules fuzzyRuleSet.

In order to improve the control accuracy of the numerical control device, the input coefficient and output coefficient in fuzzyInSet and fuzzyOutSet select larger parameters. The coefficient selects the smaller parameter. Therefore, when selecting the input and output coefficients, it is necessary to compromise between the accuracy and stability of the device. Accordingly, in this embodiment, the input coefficient and the output coefficient are set to 1.5 and 4, respectively.

The input data operation (inputError()) reads the setting value given by the motion planning from the axis setting (AxisSetpoint), and reads the actual position value of the current axis from the axis detection AxisSense to form the error e and error change rate ec of the fuzzy PID control.

The error judgment operation (ctrlAccuracy()) evaluates the execution of the current PID control operation; if the current control error is small, that is, if the value of the input error e is less than the control accuracy of the system by an order of magnitude, it means that the parameter value of the current PID control operation is relatively low. OK, no adjustment is needed, skip the fuzzy self-tuning related modules at this time, and directly run the PID control operation (that is, the PID regulator); otherwise run the fuzzy self-tuning related modules to adjust the relevant parameters of the PID regulator;

The fuzzy operation (fuzzy()) maps the input value from the input data operation into a fuzzy value in a certain interval according to the fuzzy method and the fuzzy membership function, and forms a language variable as the input parameter of the fuzzy reasoning operation. Based on the speed and accuracy requirements of the numerical control device, the error e and the error change rate ec are used as the domain of discourse on the fuzzy set; where:

e={-0.75, -0.5, -0.25, 0, 0.25, 0.5, 0.75}

ec={-0.15, -0.1, -0.05, 0, 0.05, 0.1, 0.15}

The corresponding fuzzy subset is

e, ec={LN (the negative direction is too large), MN (the negative direction is centered), SN (the negative direction is small), ZZ (zero), SP (the positive direction is small), MP (the positive direction is centered), LP ( The positive direction is too large)}

In order to enable the device to respond in real time to changes in the error e and the error change ec in a large range, the membership function of the error e and the error change ec is defined as an open interval, and the membership function of the interval boundary is a trapezoidal membership function. Others are triangular membership functions.

The fuzzy inference operation (fuzzyInf()) derives the parameters K _p , K _i , K _d corresponding to the input error e and the error change ec according to the fuzzy inference rules and the fuzzy inference method, and the output value is in the form of a fuzzy value;

The induction of the proportional, integral and differential parameters K _p , K _i , and K _d derivation rules: Based on a large number of engineering experiments, the fuzzy reasoning method selects the product reasoning machine, and the product reasoning machine uses the independent reasoning principle of fuzzy combination and Mamdani Product meaning, all t-norm operators use the algebraic product operator, and all s-norm operators use the maximum operator.

The specific derivation rules of this embodiment are shown in Table 1, Table 2, and Table 3. It can be seen from the rule table that the axis control rules of the CNC device are independent of each other.

Table 1 Fuzzy rule table of K _p

Table 2 Fuzzy rule table of K _i

Table 3 Fuzzy rule table of K _d

The defuzzification operation (deFuzzy()) defuzzifies the output value of the fuzzy inference operation according to the output membership function, and obtains the proportional, integral, and differential adjustment gain parameters (Δp, Δi, Δd) of the PID control operation; in order to make the solution The calculation process of fuzzification is simple, intuitive and reasonable, in order to improve the running speed and control precision of the device, the center average defuzzifier is selected as the defuzzifier.

The determination of the output membership function is related to the initial value setting of the PID regulator parameters and the variation of the PID regulator parameters in each control cycle. The smaller the difference between the initial value setting and the steady-state value of the PID control operating parameters, the smaller the amount of change that needs to be adjusted, and the smaller the range of the membership function interval. If the PID control operating parameter variation is large in a single cycle, then the execution efficiency of the system (the realization device of the inventive method) is high, but the stability is poor, and the axis may be positively out of tolerance; if the PID control operating parameter variation is small in a single cycle, Then the stability of the system is good, but the execution efficiency is low, and the axis negative tolerance may occur. According to the above characteristics of the device, the change interval of the proportion Δp is selected as [-1, +1], the change interval of the integral ΔI is [-0.1, +0.1], and the change interval of the differential Δd is [-0.2, +0.2].

The gain control operation (gainCtrl()) adjusts the proportional, integral, and differential parameters K _p , K i , and K d of the PID control operation according to the value of the proportional, integral, and differential adjustment gain parameters Δp, _Δi , and _Δd obtained by the defuzzification operation value, the adjustment principle is:

{e _j +ec _j } _p ＝p _a ×Δp

{e _j +ec _j } _i ＝i _a ×Δi (3)

{e _j +ec _j } _d =d _a ×Δd

Where p _a , i _a , d _a are real numbers greater than zero, and j is an ordinal number; its function is to increase or decrease the speed of fuzzy self-tuning. If the values of p _a , _ia , and d _a are large, the setting time of the PID control operation parameters will be shortened; if the values of p _a , ia , _and d _a are small, a more stable and smooth control effect can be obtained.

The control protection operation (ctrlSafe()) sets the upper limit of the adjustment of the proportional, integral and differential parameters of the PID control operation to prevent serious consequences from overshooting for some reason. During the machining process of the CNC device, if the feed exceeds the actual load capacity of the tool, not only will the machining task not be completed, but it may cause tool damage and even casualties. Therefore, it is necessary to set the range of PID control operation parameters. The variation range of the proportional, integral and differential parameters of the PID control operation: if the proportional parameter K _p of the PID control operation is greater than the set maximum value p _max , then the value of the proportional parameter K _p of the control operation is equal to the set maximum value p _max ; if the integral parameter K _i of the PID control operation is greater than the set maximum value i _max , then make the value of the integral parameter K _i of the PID control operation equal to the set maximum value i _max ; if the differential parameter K of the PID control operation _d is greater than the set maximum value d _max , then the value of the differential parameter K _d of the PID control operation is equal to the set maximum value d _max ; among them, p _max , i _max , and d _max are the proportion and integral of the PID control operation, respectively , The upper limit of differential parameter adjustment, the specific value is set according to the actual performance index of the axis, tool and motor;

The PID control operation (PIDCtrl()) adopts the traditional PID control method to realize the control function of the axis according to the values of the proportional, integral and differential parameters K _p , K _i , and K _d obtained from the control and protection operation; Modulo conversion, converted into a voltage value, to achieve the control function of the servo axis.

As shown in Figure 8, the fuzzy PID control operation is to call the above input data, error judgment, fuzzification, fuzzy reasoning, defuzzification, gain control, control protection and PID control operation to realize the optimal control of the servo axis of the machine tool; the specific process It is as follows: firstly perform the input data operation, and then perform the error judgment operation after the input data operation, if there is an error, perform the fuzzification operation, fuzzy reasoning operation, defuzzification operation, gain control operation, and then enter the control and protection operation; if there is no error, then Go directly to the control and protection operation; finally perform the PID control operation.

During the operation of the realization device of the above-mentioned soft and hard structure, the user can input the workpiece program through the human-machine interface. Under the control of task coordination, the workpiece program is interpreted by the interpreter to form the machine tool control command, and the machine tool control command is generated by motion planning. After setting the value, enter the control phase of the servo axis. The shaft control using the fuzzy PID method can adjust the control parameters online according to the motion state of the shaft to form an intelligent decision-making process for the device. For the decision-making process, refer to the specific flow of the fuzzy PID control operation shown in Figure 8. The shaft control takes the sampling cycle of the device realized by the present invention as the operation cycle. First, the actual position of the shaft is detected through the encoder interface circuit. The actual position value and the set value generate an error e and an error change rate ec, and e and ec are fuzzy through fuzzy self-tuning. Linguistic variables are formed through fuzzy reasoning and defuzzification, and the proportional, integral, and differential parameters of the PID regulator are generated through fuzzy reasoning and defuzzification. Under the control of the parameters, the PID regulator controls the movement of the servo axis of the machine tool to complete a servo cycle decision-making and control process.

(3) Execution effect of the present invention:

The control object adopts Yaskawa Σ-2 servo and motor commonly used in numerical control devices, and its main parameters are as follows:

·Maximum speed = 2000 rev/s

·Maximum processing speed = 10 m/s

·Maximum rated voltage = 10V

· Mechanical pitch = 6 mm

The main parameters of the inventive method and the intelligent realization device used are as follows:

· Interpolation period = 0.002 seconds

·Axis maximum speed = 10.00 m/s

· Following error = 1.200mm

·Minimum following error＝0.010mm

· Servo cycle = 0.0005 seconds

·p=70 (initial value of PID control parameter K _p )

·i=0.1 (initial value of PID control parameter K _i )

·d＝0.05 (initial value of PID control parameter K _d )

Based on the above parameters, adopt the present invention, the servo axis error change trajectory is as shown in Figure 9 (the abscissa represents the sampling period, and the ordinate represents the error value in the current sampling period), and adopts the traditional PID control method, the servo axis error change trajectory As shown in Figure 10. The following conclusions can be drawn from the comparison of the servo axis error change trajectory: the realization device of the fuzzy PID control method

1. The realization device of the fuzzy PID control method has high execution efficiency. The present invention realizes the control target within about 80 sampling periods, while the PID control method needs more than 100 sampling periods to reach the control target;

2. The realization of fuzzy PID control method The device has high control precision and smooth control curve. In the control process of the present invention, there is no large jitter phenomenon in the error change. However, there is error jitter in the PID control process.

3. The realization of fuzzy PID control method has good dynamic performance of the device. Because for the same initialization control parameters, the method implementing device of the present invention can adjust the value of the PID regulator parameter according to the actual operating conditions on site during the control process, so it is more robust than the PID control method.

Claims (8)

1. A fuzzy PID control method for a numerically controlled machine tool, which adopts man-machine interface, task coordination, motion control and programmable controller, and loads motion control and programmable controller into the kernel space of the system, and man-machine interface and task coordination are partially loaded In the user space of the system, they communicate with each other through the shared communication buffer of the system; the motion control includes two parts: motion planning and axis control; it is characterized in that: the axis control is dynamically determined by the fuzzy PID control method according to the set value of the axis motion planning The control parameters of the axis and realize the stable control of the axis, thus forming the movement of the servo axis in the process of CNC machine tool processing; The fuzzy PID control method is composed of two parts, a parameter-adjustable PID regulator and a fuzzy self-tuning mechanism, wherein: the PID regulator is used as the basic control unit, aiming at the nonlinear characteristics of the machine tool servo axis during processing and various processing Requirements, introduce fuzzy control technology to realize the adjustment of the regulator parameters; the fuzzy self-tuning mechanism receives the set value of the motion planning output and detects the actual position value of the shaft through the encoder to form the error e and the error change rate ec, and then the fuzzy self-tuning The fuzzification in the mechanism is processed to form language variables. After fuzzy reasoning and defuzzification processing, the parameters of the PID regulator are generated. Under the control of the parameters, the PID regulator controls the movement of the servo axis of the machine tool, thereby completing the servo axis. control process. 2. by the described fuzzy PID control method of numerically controlled machine tool of claim 1, it is characterized in that: described fuzzy inference is to take error e and error rate of change ec as input, and reasoning rule form is as follows: If(e is...)and(ec is...)then(K _p is...)(K _i is...)(K _d is...) (1) Based on the rules, it is concluded that the PID parameters of the PID regulator are nonlinear functions of the error e and the error change ec, which can be specifically recorded as: ${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; p</span>}}$ ${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$ ${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}}^{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; \{</span>{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; ec</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \}</span>}_{\mathrm{<span\; class="notranslate">\; d</span>}}$ in,

Respectively represent the initial values of the proportion, integral and differential in the PID parameters, {e _j +ec _j } _p , {e _j +ec _j } _i , {e _j +ec _j } _d represent the obtained after fuzzy reasoning and defuzzification Adjust the proportional, integral and differential variation of PID parameters; where j is an ordinal number. 3. a kind of realization device by the said fuzzy PID control method of numerical control machine tool of claim 1, it is characterized in that: this device is based on ISA bus, is powered by power supply (4), and it comprises: display (1), central controller (2 ), the machine operation panel (3), the axis control circuit (5) and the interface circuit (6), wherein: the central controller (2) communicates with the display (1) and the machine operation panel (3), and communicates with the axis through the ISA bus The control circuit (5) communicates with the interface circuit (6), and the control program is installed in the FLASH or electronic disk of the central controller (2), realizing the fuzzy PID control of the CNC machine tool; The control program includes axis control, axis characteristics, axis constraints, axis setting, axis detection, axis drive, and axis fuzzy PID control parts; the axis control is the main program, wherein: the axis characteristics record the dynamic characteristics and static characteristics of the axis; Constraint records the movement ability of the axis; the axis setting is used as the interface of fuzzy PID control and motion planning, and records the axis setting position; the axis detection is used as the interface of the position encoder to detect the actual running position and speed of the axis; the axis drive is used as the interface of the servo axis, Provide the signal of motor operation; axis fuzzy PID control uses the data of axis characteristics, axis constraints, axis detection, and axis drive according to the fuzzy PID control method to realize the optimal control of axis motion. 4. by the realization device of the described CNC machine tool fuzzy PID control method of claim 3, it is characterized in that: described fuzzy PID control comprises: initialization operation, input data operation, error judgment operation, fuzzification operation, fuzzy reasoning operation, defuzzification Transformation operation, gain control operation, control protection operation, PID control operation, fuzzy PID control operation, where: Initialization operations, specifically including PID control operation ratio, integral, differential parameter initialization and fuzzy self-tuning parameter initialization; The input data operation reads the setting value given by the motion planning from the axis setting, and reads the actual position value of the current axis from the axis detection to form the error e and the error change rate ec of the fuzzy PID control; The error judgment operation evaluates the execution of the current PID control operation; if the value of the input error e is less than a unit order of magnitude of the control accuracy of the system, no adjustment is required, and the fuzzy control related modules are skipped at this time, and the PID control operation is directly run; otherwise, the fuzzy control operation is run Control related modules and adjust related parameters of PID control; The fuzzy operation maps the input value from the input data operation into a fuzzy value on a certain interval according to the fuzzy method and the fuzzy membership function. The domain of the fuzzy value is: e={-0.75, -0.5, -0.25, 0, 0.25, 0.5, 0.75} ec={-0.15, -0.1, -0.05, 0, 0.05, 0.1, 0.15} The corresponding fuzzy subset is: e, ec={LN (the negative direction is too large), MN (the negative direction is centered), SN (the negative direction is small), ZZ (zero), SP (the positive direction is small), MP (the positive direction is centered), LP ( The positive direction is too large)} The fuzzy inference operation deduces the parameters K _p , K _i , K _d corresponding to the input error e and the error change ec according to the fuzzy inference rules and fuzzy inference method, and the output value is in the form of fuzzy value; The defuzzification operation defuzzifies the output value of the fuzzy reasoning operation according to the output membership function, and obtains the proportional, integral and differential adjustment gain parameters △p, △i, △d of the PID control operation; The gain control operation adjusts the proportional, integral and differential parameters K _p , K i , and K d of the PID control operation according to the values of the proportional, integral, and differential adjustment gain parameters △p, △ _i , and △ _d obtained by the defuzzification operation , the adjustment principle is: {e _j +ec _j } _p ＝p _a ×△p {e _j +ec _j } _i ＝i _a ×△i(3) {e _j +ec _j } _d ＝d _a ×△d Among them, p _a , i _a , d _a are real numbers greater than zero, and j is an ordinal number; The control protection operation sets the upper limit of the proportional, integral and differential parameter adjustment of the PID control operation, and the variation range of the proportional, integral and differential parameters of the PID control operation: if the proportional parameter K _p of the PID control operation is greater than the set maximum value P _max , then the value of the proportional parameter K _p of the control operation is equal to the set maximum value p _max ; if the integral parameter K _i of the PID control operation is greater than the set maximum value i _max , then the integral parameter K _i of the PID control operation is set The value of is equal to the set maximum value i _max ; if the differential parameter K _d of the PID control operation is greater than the set maximum value d _max , then the value of the differential parameter K _d of the PID control operation is equal to the set maximum value d _max ; Among them, p _max , i _max , and d _max are the upper limits of the proportional, integral, and differential parameter adjustments of the PID control operation, and their specific values are set according to the actual performance indicators of the shaft, tool, and motor; PID control operation According to the value of proportional, integral and differential parameters K _p , K _i , K _d obtained from the control and protection operation, the traditional PID control method is adopted to realize the control function of the axis; The fuzzy PID control operation is to call the above input data, error judgment, fuzzification, fuzzy reasoning, defuzzification, gain control, control protection and PID control operation to realize the optimal control of the servo axis of the machine tool; the specific process is: firstly, input data Operation, after the input data operation, the error judgment operation is performed. If there is an error, the fuzzification operation, fuzzy reasoning operation, defuzzification operation, gain control operation are performed in sequence, and then enter the control protection operation; if there is no error, go directly to the control protection operation; Finally, the PID control operation is performed. 5. by the realization device of the described numerically controlled machine tool fuzzy PID control method of claim 3, it is characterized in that: the ratio of PID control operation, integral, the initialisation of differential parameter comprise the proportion of PID control operation, integral, initialisation of initial value of differential parameter and PID The initialization of the maximum value of the proportional, integral and differential parameters of the control operation; the initialization of fuzzy self-tuning parameters includes the initialization of the fuzzy control input and output membership function, the initialization of fuzzy control input and output coefficients, and the initialization of fuzzy inference rules. 6. by the realization device of the described CNC machine tool fuzzy PID control method of claim 3, it is characterized in that: the degree of membership function of the domain error e and error variation ec of definition fuzzy value is an open interval, and the degree of membership function of interval boundary selects trapezoid membership function, others are triangular membership functions. 7. by the realization device of the fuzzy PID control method of the numerical control machine tool described in claim 4, it is characterized in that: wherein said proportion, integral, differential parameter K _p , K _i , K _d derivation rule is: fuzzy reasoning method selects the product reasoning machine , the product reasoning machine uses the independent inference principle of fuzzy combination, adopts Mamdani product meaning, all t-norm operators use algebraic product operator, and all s-norm operators use maximum operator. 8. The realization device of the fuzzy PID control method for a numerically controlled machine tool according to claim 4, characterized in that: said defuzzification operation selects a center average defuzzifier as the defuzzifier.

Images