Servo hydraulic clamping control method and system based on fuzzy backoff time domain control

Abstract

The present invention discloses a servo hydraulic clamping control method and system based on fuzzy backward time domain control; the present invention relates to the field of numerical control electromechanical technology. A fuzzy controller receives the control axis variable X(t) transmitted by the CNC control system at the current time step t; a membership function is used to convert the control axis variable X(t)' into a fuzzy variable. The membership of the displacement error e1 and the displacement error change rate e2 is calculated; the fuzzy set of the displacement error e1 and the displacement error change rate e2 is divided into three fuzzy subsets: negative N, zero Z and positive P; the present invention introduces a fuzzy controller, and utilizes the membership function and the fuzzy rule base to achieve accurate modeling and intelligent control of the state of the hydraulic clamping mechanism. Fuzzy control effectively overcomes the limitations of traditional control methods by fuzzifying precise input parameters, performing reasoning using fuzzy logic, and then defuzzifying to obtain precise control signals.

Description

Technical Field

[0001] The present invention relates to the field of CNC electromechanical technology, and in particular to a control method for a hydraulic clamping system based on data drive, which belongs to a type of CNC machine tool control technology. The present invention also relates to predictive control (MPC) based on a data-driven model, and in particular to a servo hydraulic clamping control method and system based on fuzzy backoff time domain control. Background Art

[0002] CNC (Computerized Numerical Control) stands for computerized numerical control. A CNC machine tool is an automated machine tool equipped with a program control system. This control system logically processes and decodes programs specified by control codes or other symbolic instructions, thereby enabling the machine tool to operate and process parts. During CNC machining, workpieces must be clamped and released, a function accomplished by a hydraulic release mechanism. The hydraulic system transmits pressure through fluid to drive the clamping and releasing mechanisms to perform various operations. In CNC machining, the hydraulic release mechanism typically consists of a hydraulic cylinder, control valve, oil circuit system, and fixture body. When pressurized oil enters the hydraulic cylinder, the piston is pushed outward by the pressure, driving the fixture body to clamp the workpiece. Conversely, when the pressurized oil is discharged from the hydraulic cylinder, the piston returns to its original position under the action of spring force or its own weight, and the fixture body releases.

[0003] In theory, the hydraulic release mechanism works by introducing hydraulic oil at a certain pressure into the cylinder through the oil inlet, causing the piston to move downward, pressing the rotary table's annular brake to achieve work clamping. The hydraulic oil is then depressurized, and under the action of the spring, the piston retracts upward, releasing the rotary table's annular brake and releasing the worktable. In practice, however, errors in the hydraulic release mechanism arise from the following reasons: Because the hydraulic mechanism's response speed is relatively slow, there is a time delay during the clamping and release processes; this further makes it impossible for the piston to be completely synchronized during the clamping and release processes, resulting in slight angular deflections; based on this, errors begin to accumulate; the angular error generated during each release is very small; as the number of releases increases, these tiny angular errors gradually accumulate, eventually causing the angular positioning accuracy to exceed the allowable range.

[0004] Generally speaking, the existing technical solutions generally delay 500ms before opening the hydraulic valve after receiving the clamping signal and detecting the axis in position signal; after detecting the pressure relay signal, delay 200ms before giving the axis prohibition signal and turning off the axis enable; avoid interrupting the precise positioning of the axis and leave enough response time for the hydraulic clamping mechanism. When the release command is detected, the axis prohibition is first canceled, the axis enable is loaded, and the valve is turned off after a delay of 200ms; when the pressure relay signal is detected to be 0, the axis rotation command is loaded after another delay of 500ms. Loading the axis enable in advance reserves time for the mechanical displacement to eliminate errors; ensuring that the clamping mechanism is completely loosened and in place to avoid friction between the brake disc and the piston. But this creates a new problem: when the control axis is unable to perform position control (mainly in the servo-off state, such as when the mechanical handwheel controls the movement of the worktable, or when the mechanism to prevent the servo motor from overcurrent is turned on), if the axis generates mechanical movement, the mechanical displacement feedback pulses will be accumulated in the error counter, which will cause the CNC machine tool to reflect the mechanical movement error count to the current position, which will cause the accumulation of errors during relative instruction movement; if the CNC machine tool uses relative instructions to move or rotate the axis and perform position tracking at this time, the errors generated by each servo shutdown will accumulate and then continue to increase until the positioning accuracy exceeds the tolerance; at this time, the machine can only be shut down for a long time and its main components disassembled, and then manually fine-tuned using a large number of precision testing tools to eliminate the errors.

[0005] Therefore, many existing technologies attempt to solve the problem that the precision error is difficult to adjust when the control axis cannot perform position control, for example:

[0006] (1) The document "Yin Zuosheng, Wang Chao, Wang Wei, et al. Compensation method for the angle error of the rotary axis of hydraulic clamping [J]. Mechanical and Electrical Product Development and Innovation, 2024, 37(02): 186-188+192." discloses a solution (see section 2.2.2 for details) that solves the technical problem by manually setting not to perform position tracking, modifying electrical signals, and controlling PLC timing. However, it is highly subjective and has a low degree of automation, making it difficult to adapt to efficient processing scenarios;

[0007] (2) The document "Zhu Xinglong. Analysis of Stability and Steady-state Error of a Hydraulic Servo Joint [J]. Control Engineering, 2006.13(1):87-90" discloses a method for compensating for stability and steady-state error, which mainly solves this problem by establishing a dynamic model. However, this model is mainly applicable to multi-axis robots such as robotic arms. For hydraulic clamping systems such as CNC machine tools, which generally have fixed spatial motion and trajectory, its model architecture is overly redundant and unnecessary, and will increase the computational burden.

[0008] (3) The document "Zhang Shaoxin. Dynamic Error Analysis and Compensation Methods of FANUC-0i CNC Machine Tool Servo System [J]. Journal of Anhui Institute of Architecture and Technology: Natural Science Edition, 2013, 21(3):4" discloses a data-driven architecture that uses HRV control, combined with speed feedforward control, acceleration / deceleration time constant adjustment, friction, and deceleration methods to achieve compensation. However, this solution requires human intervention and subjective judgment on whether to execute it. Although it has some automation functions, it cannot actually achieve the effect of predictive control (MPC) based on data-driven models required in today's CNC field. At the same time, it is difficult to effectively handle nonlinear and uncertain factors in hydraulic systems. For example, the piston position and pressure of the hydraulic cylinder are not simply linearly related to the error.

[0009] To this end, the present invention proposes a servo hydraulic clamping control method and system based on fuzzy backoff time domain control. Summary of the Invention

[0010] In view of this, the present invention aims to provide a servo-hydraulic clamping control method and system based on fuzzy backoff time domain control to solve or alleviate the technical problems existing in the prior art, namely, how to enable the hydraulic clamping mechanism to pre-determine whether error accumulation has occurred, handle nonlinear and uncertain factors, and automatically modify the electrical signal and PLC timing to achieve an automated data-driven process with autonomous detection and autonomous adjustment, and at least provide a beneficial option for this purpose. The technical solution of the present invention is achieved as follows:

[0011] First, the servo hydraulic clamping control method based on fuzzy backward time domain control:

[0012] (I) Overview:

[0013] The present invention addresses the response delay and synchronization error accumulation problems of the hydraulic unclamping mechanism of CNC machine tools during the clamping and releasing processes. By real-time monitoring of the hydraulic cylinder piston position and pressure parameters, and using a fuzzy controller to perform fuzzification, fuzzy reasoning and defuzzification processing, a precise control signal is generated to adjust the opening and closing timing of the hydraulic valve. This solution introduces a backward time domain control mechanism that can evaluate the control effect based on the real-time feedback signal and dynamically adjust the control strategy when errors occur to correct future control signals. Through this automated data-driven process of self-detection and self-adjustment, this technical solution can effectively pre-identify and reduce error accumulation, significantly improve the positioning accuracy and stability of the hydraulic unclamping mechanism of CNC machine tools, thereby improving overall processing accuracy and efficiency, and providing strong support for the efficient and precise processing of CNC machine tools.

[0014] (2) Technical solution:

[0015] To achieve the above goals, after the sensor monitors the piston position l and pressure p parameters of the hydraulic cylinder in real time, the collected data is converted into digital signals, transmitted to the fuzzy controller, and the following steps S1 to S6 are executed.

[0016] 2.1 Step S1, the fuzzy controller performs fuzzification operations:

[0017] The fuzzy controller receives the control axis variable X(t)=[e1,e2] transmitted by the CNC control system at the current time step t, where e1 is the displacement error and e2 is the rate of change of the displacement error; the control axis variable X(t) is converted into the fuzzy variable X(t)' using the membership function.

[0018] 2.1.1 Step S100, calculate the membership degree:

[0019] Calculate the membership of the displacement error e1 and the displacement error change rate e2; assume that the fuzzy set of the displacement error e1 and the displacement error change rate e2 is divided into three fuzzy subsets: negative N, zero Z and positive P:

[0020] (1) Gaussian membership function of displacement error e1:

[0021] Membership function μ of negative N fuzzy subsets N (e1): ;

[0022] Membership function μ of zero-Z fuzzy subset Z (e1): ;

[0023] Membership function μ of positive P fuzzy subset P (e1): ;

[0024] Among them, c1, c2, c3 are the centers of negative, zero, and positive fuzzy subsets, respectively, and σ1, σ2, and σ3 are their standard deviations, respectively.

[0025] (2) Gaussian membership function of displacement error change rate e2:

[0026] Membership function μ of negative N fuzzy subsets N (e2): ;

[0027] Membership function μ of zero-Z fuzzy subset Z (e2): ;

[0028] Membership function μ of positive P fuzzy subset P (e2): ;

[0029] Among them, c4, c5, c6 are the centers of negative, zero, and positive fuzzy subsets, respectively, and σ4, σ5, and σ6 are their standard deviations, respectively.

[0030] 2.1.2 Step S101, fuzzification operation:

[0031] The membership degrees calculated in step S100 are combined to form the fuzzy variable X(t)':

[0032] ;

[0033] Among them, [·,·] represents vector concatenation operation.

[0034] 2.2 Step S2, the fuzzy controller performs fuzzy reasoning:

[0035] The fuzzy controller performs fuzzy reasoning based on the fuzzy variable X(t)' and the rule base R at the current time step t, deduces the possible values ​​of the output domain, and obtains the fuzzy output Fo as the control quantity required for the opening of the hydraulic valve;

[0036] 2.2.1 Step S200, rule evaluation (matching):

[0037] For each rule r in the rule base R i , execute the following "IF-THEN" conditional statement rule matching: "If yes and yes Then Y is ”;

[0038] Among them, X1' and X2' are components of the fuzzy variable X'(t) (two-dimensional fuzzy controller), is the input fuzzy set, B i is the output fuzzy set.

[0039] 2.2.1.1 Rule Base R:

[0040] The input variable X1′ is the displacement error e1, and the input variable X2′ is the displacement error change rate e2. The output variable Y is the hydraulic valve opening control value F o .

[0041] 2.2.2 Step S201, calculate the matching degree (activation degree) μ ​​of each rule ri :

[0042] ;

[0043] in, and X1' and X2' belong to the fuzzy set and The degree of membership.

[0044] 2.2.3 Step S202, polymerization:

[0045] For each output fuzzy set B i , calculate its aggregated membership function :

[0046] ;

[0047] in, The output value y belongs to the fuzzy set B i The membership degree of y i is the possible value of the fuzzy output belonging to the maximum possible case;

[0048] 2.2.4 Step S203: Execute push message:

[0049] Use the DS evidence theory algorithm to output the best possible value y in the domain i '.

[0050] S2030, define Basic Probability Assignment (BPA):

[0051] For every possible output value y i , define the basic probability assignment m(y i ),satisfy and ; where Θ is the set of all possible values ​​of the output universe.

[0052] S2031, combine evidence using Dempster's combination principle:

[0053] For the i-th and j-th fuzzy output possible values ​​y i and y j , take the two as evidence sources, namely m1 and m2, then the basic probability of the combination is assigned to m 1,2 (y i ) for:

[0054] ;

[0055] Where K is a normalization constant, defined as: ;

[0056] S2032, calculate the trust function Bel(y i ): ;

[0057] S2033, calculate the likelihood function Pl(yi ): ;

[0058] S2034, decision and assignment: Based on the trust function and likelihood function, use the combination of trust and uncertainty to calculate each possible output value y i Overall rating:

[0059] Select the y with the highest score i (Even if Bel(y i ) maximum) as the best possible value y i ':

[0060] ;

[0061] 2.2.5 Step S204, defuzzification:

[0062] For the discrete case: ;

[0063] Among them, Fo is the fuzzy output, marking is its membership function, and x is the possible value range of the fuzzy output variable.

[0064] The fuzzy output Y' can also be converted into the control quantity F o : ;

[0065] 2.3 Step S3, the fuzzy controller performs electrical signal conversion:

[0066] The fuzzy output Fo is defuzzified using the center of gravity method and then converted into a control signal e(t), which is then transmitted to the PLC controller: ;

[0067] Among them, the control signal e(t) is an electrical signal that adjusts the opening and closing time ratio or frequency of the hydraulic valve; the integration is performed over the entire definition domain of the fuzzy output variable.

[0068] 2.4 Step S4, timing control:

[0069] The PLC controller controls the opening and closing timing of the hydraulic valve according to the received control signal e(t), thereby achieving precise control of the hydraulic cylinder and completing the clamping or loosening of the workpiece.

[0070] Based on the preset complete opening and closing cycle T of the hydraulic valve, the hydraulic valve remains open for a period of time T within one cycle. on And the closing time T within a cycle off Calculated as: ;

[0071] Among them, the control signal e(t) changes between 0 and 1, indicating the opening ratio.

[0072] 2.5 Step S5, perform backoff time domain:

[0073] That is, the process of using past and current information to predict and adjust future control signals: the sensor monitors the status of the hydraulic cylinder in real time and transmits the feedback signal f(t) consisting of the new piston position l' and the new pressure p' to the fuzzy controller; the fuzzy controller evaluates the control effect based on the feedback signal f(t); if there is an error, the back-off time domain is executed, and the fuzzy controller adjusts the control strategy to correct the error in the next time step t+1.

[0074] S500, assume that the desired piston position corresponding to the opening of the hydraulic valve controlled in step S4 is l d and the expected pressure is p d ; then calculate the position error e l (t) and pressure error e p (t):

[0075] e l (t) = l'-l d ;

[0076] e p (t) = p'-p d ;

[0077] S501, define the error vector E(t) = [e l (t), e p (t)]; where e l (t) and e p (t) are the errors of position and pressure, respectively;

[0078] S502, the fuzzy controller performs the following operations according to the error vector E(t):

[0079] S5020, fuzzification: calculate the membership degree of the error E(t) according to the form of steps S100-S101 and convert it into a fuzzy set;

[0080] S5021, rule evaluation: according to the fuzzy logic rule base R, the fuzzy output is evaluated in the form of steps S200 to S204;

[0081] S5022, defuzzification: according to the form of step S4, the fuzzy output is converted into a new control signal e(t+1); the control signal e(t+1) is updated for the next time step t+1 to correct the error.

[0082] 2.6 Step S6, loop execution:

[0083] Steps S1 to S5 are executed cyclically to realize autonomous detection and autonomous adjustment of the hydraulic clamping mechanism until the workpiece is processed and released and removed.

[0084] (3) Mechanism for resolving technical issues:

[0085] 3.1 Fuzzy processing:

[0086] The fuzzy controller receives the control axis variable X transmitted by the CNC control system at the current time step t; it uses the membership function to convert the control axis variable X(t) into a fuzzy variable X(t)′. This process allows the precise input value to be converted into a fuzzy set, facilitating subsequent fuzzy reasoning.

[0087] The fuzzy controller performs fuzzy inference based on the fuzzy variable X(t)′ at the current time step t and the rule base R. The rule base R contains a series of "if-then" conditional statements used to infer the output fuzzy variable (i.e., the hydraulic valve opening control variable Fo) based on the input fuzzy variables. By calculating the matching degree (activation degree) and the aggregated membership function of each rule, the DS evidence theory algorithm is used to output the best possible value in the domain.

[0088] 3.2 Defuzzification and electrical signal conversion:

[0089] The fuzzy output Fo is defuzzified using the center of gravity method and converted into a precise control signal e(t). The control signal e(t) is transmitted to the PLC controller to adjust the opening and closing time ratio or frequency of the hydraulic valve.

[0090] The PLC controller controls the opening and closing timing of the hydraulic valve based on the control signal e(t) it receives, achieving precise control of the hydraulic cylinder. By adjusting the opening and closing time of the hydraulic valve, the workpiece can be clamped or released.

[0091] 3.3 Real-time feedback and back-off time domain control:

[0092] The sensor monitors the new state of the hydraulic cylinder in real time, generating a feedback signal f(t) that is transmitted to the fuzzy controller. The fuzzy controller evaluates the control effectiveness based on the feedback signal and calculates the position and pressure error vectors. If an error exists, the fuzzy controller performs back-off control, using past and current information to predict and adjust future control signals. The error rate of change is calculated, and the fuzzy input is updated (including the current error and the error rate of change). Fuzzification, rule evaluation, and defuzzification are repeated to obtain a new control signal e(t+1). The updated control signal e(t+1) is used at the next time step t+1 to correct the error.

[0093] 3.4 Cyclic execution and autonomous adjustment:

[0094] The fuzzy controller loops through these steps, enabling autonomous detection and adjustment of the hydraulic release mechanism. Through continuous monitoring, reasoning, and adjustment, this technical solution can proactively identify potential error accumulation and automatically modify the electrical signal and PLC timing, thereby improving the positioning accuracy and stability of the CNC machine tool's hydraulic release mechanism.

[0095] Second, the servo hydraulic clamping control system based on fuzzy backoff time domain control:

[0096] The system is used to implement the servo hydraulic clamping control method as described above, including:

[0097] A fuzzy controller for receiving the control axis variable X transmitted by the CNC control system and performing fuzzification operations, fuzzy reasoning and electrical signal conversion;

[0098] A PLC controller for receiving the control signal e(t) calculated by the fuzzy controller and controlling the opening and closing timing of the hydraulic valve;

[0099] A sensor unit mounted on the piston rod, used to feed back the detected control axis variable X to the CNC control system and to feed back the feedback signal f(t) to the fuzzy controller;

[0100] A memory storing program instructions and a controller connected to the memory, when the controller executes the program instructions, causes the fuzzy controller to execute the servo hydraulic clamping control method as described above.

[0101] Compared with the prior art, the present invention has the following beneficial effects:

[0102] 1. Application and optimization of fuzzy control: Traditional control methods are difficult to effectively deal with nonlinear and uncertain factors in hydraulic systems. The present invention introduces a fuzzy controller, and uses membership functions and fuzzy rule bases to achieve accurate modeling and intelligent control of the state of the hydraulic clamping mechanism. Fuzzy control effectively overcomes the limitations of traditional control methods by fuzzifying precise input parameters, using fuzzy logic for reasoning, and then defuzzifying to obtain precise control signals. The defuzzification method and DS evidence theory algorithm adopted ensure the smooth conversion of fuzzy output to precise control signals, further improving the accuracy and reliability of the control signals. This data-driven automated adjustment process not only improves the system's adaptability, but also reduces the need for manual intervention, making the hydraulic clamping control of CNC machine tools more intelligent and efficient.

[0103] 2. Introduction of a Backward Time Domain Control Mechanism: Traditional control methods often make decisions based on the current state and lack the ability to predict and adjust future states. This invention introduces a backward time domain control mechanism, leveraging past and current information to predict and adjust future control signals. By monitoring errors in real time and calculating the error rate of change, the fuzzy controller can proactively identify the trend of error accumulation and adjust the control strategy in advance for the next time step. This significantly reduces error accumulation, improving system stability and control accuracy, particularly during long-term operation or during frequent clamping and loosening.

[0104] 3. Data-Driven and Automated Adjustment Process: Traditional control methods often require manual intervention to adjust control parameters, lacking autonomy and intelligence. This invention implements an automated, data-driven process with autonomous detection and adjustment. Through real-time monitoring and data analysis, the system automatically determines whether control strategies need adjustment and dynamically generates new control signals. This improves the system's adaptability and intelligence, reduces the need for manual intervention, lowers maintenance costs, and improves production efficiency.

[0105] Fourth, it indirectly improves machining accuracy and stability: By precisely controlling the clamping and releasing processes of the hydraulic release mechanism, the accumulated errors are reduced, improving the machining accuracy and stability of CNC machine tools. This precise control and reduction of accumulated errors reduces wear on the hydraulic system and mechanical components, thereby extending the life of the equipment. The automated and intelligent control process reduces manual intervention and downtime, improving production efficiency. Furthermore, higher machining accuracy and stability reduce scrap rates and enhance economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0107] Figure 1 Schematic diagram of the method flow of the present invention;

[0108] Figure 2 Schematic diagram of the execution architecture of steps S1 to S2 of the present invention;

[0109] Figure 3 This is a schematic diagram of the push effect of the present invention;

[0110] Figure 4 Schematic diagram of the distribution of the independent variables corresponding to the dependent variables of the present invention;

[0111] Figure 5 Schematic diagram of the cycle of step S6 of the present invention;

[0112] Figure 6 Schematic diagram for comparison of test examples of the present invention;

[0113] Figure 7 Schematic diagram of finite element simulation of the control group of the test example of the present invention;

[0114] Figure 8 Schematic diagram of finite element simulation of the experimental group of the test example of the present invention. DETAILED DESCRIPTION

[0115] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. The following description sets forth many specific details to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways than those described herein, and those skilled in the art can make similar improvements without violating the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0116] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. The devices disclosed in the embodiments are described briefly because they correspond to the methods disclosed in the embodiments. For relevant details, refer to the method description.

[0117] Example 1: Figure 1 As shown, this embodiment discloses a servo hydraulic clamping control method based on fuzzy backward time domain control; the method needs to be executed based on a CNC lathe and its controller as a carrier; when the CNC lathe performs processing operations, the workpiece is first clamped into its hydraulic clamping system; during normal processing operations, after receiving the clamping signal and detecting the axis in-position signal, the hydraulic valve is opened after a delay of 500ms; after detecting the pressure relay signal, the axis prohibition signal is given after a delay of 200ms to turn off the axis enable; to avoid interrupting the precise positioning of the axis and to leave sufficient response time for the hydraulic clamping mechanism. However, in the servo-off state, such as when the mechanical handwheel controls the movement of the workbench, or when the servo motor overcurrent mechanism is turned on, if the axis produces mechanical movement, the sensor monitors the piston position l and pressure p parameters of the hydraulic cylinder in real time, converts the collected data into digital signals, transmits them to the fuzzy controller, and executes the following steps S1 to S6.

[0118] In this embodiment, if Figure 2As shown in Figure 1, in step S1, the fuzzy controller performs a fuzzification operation: the fuzzy controller receives the control axis variable X(t) = [e1, e2] transmitted by the CNC control system at the current time step t, where e1 is the displacement error, representing the difference between the actual position and the desired position; e2 is the rate of change of the displacement error, reflecting the rate of change of the displacement error over time. The control axis variable X(t) is converted into a fuzzy variable X(t)' using a membership function. A membership function is a method for mapping precise values ​​onto fuzzy sets. It allows us to convert specific numerical values ​​(such as displacement error and error rate of change) into fuzzy language (such as "small," "medium," "large," etc.), thus adapting to fuzzy logic processing.

[0119] Specifically, step S100 calculates the membership degree: calculates the membership degree of the displacement error e1 and the displacement error change rate e2; assumes that the fuzzy set of the displacement error e1 and the displacement error change rate e2 is divided into three fuzzy subsets: negative N, zero Z and positive P:

[0120] (1) The Gaussian membership function of the displacement error e1 divides the fuzzy set of the displacement error e1 into three fuzzy subsets: negative N, zero Z, and positive P:

[0121] Membership function μ of negative N fuzzy subsets N (e1): ;

[0122] Membership function μ of zero-Z fuzzy subset Z (e1): ;

[0123] Membership function μ of positive P fuzzy subset P (e1): ;

[0124] Where c1, c2, and c3 are the centers of the negative, zero, and positive fuzzy subsets, respectively, representing the points where the membership function reaches its maximum. σ1, σ2, and σ3 are their standard deviations, respectively, which determine the width and shape of the membership function.

[0125] (2) The Gaussian membership function of the displacement error change rate e2 is also divided into negative N, zero Z and positive P:

[0126] Membership function μ of negative N fuzzy subsets N (e2): ;

[0127] Membership function μ of zero-Z fuzzy subset Z (e2): ;

[0128] Membership function μ of positive P fuzzy subset P (e2): ;

[0129] Among them, c4, c5, c6 are the centers of negative, zero, and positive fuzzy subsets, respectively, and σ4, σ5, and σ6 are their standard deviations, respectively.

[0130] Understandably, the Gaussian function was chosen because of its smooth, continuous shape and ability to well represent the membership distribution of fuzzy subsets. By adjusting the center and standard deviation parameters, the shape and position of the membership function can be flexibly changed to accommodate different control requirements. The displacement error and displacement error rate of change are divided into three fuzzy subsets: negative, zero, and positive. This is done to more comprehensively describe their possible value ranges and provide rich information for subsequent fuzzy reasoning. By calculating the membership, precise numerical inputs are converted into fuzzy linguistic variables, enabling the control system to more flexibly handle nonlinear and uncertain factors, thereby improving control accuracy. The introduction of the membership function makes the system more sensitive to small changes in the input variables, enabling timely adjustment of the control strategy to adapt to different processing conditions, thereby enhancing the robustness of the system. The calculated membership serves as the input for fuzzy reasoning, which is then used to make inference decisions based on the fuzzy rule base, resulting in the corresponding control output, thus realizing intelligent control of the hydraulic clamping mechanism.

[0131] Specifically, step S101 is a fuzzification operation: the membership degrees calculated in step S100 are combined to form a fuzzy variable X(t)': ;

[0132] Among them, [·,·] represents vector concatenation operation, μ e1 and μ e2 are the membership vectors of displacement error e1 and displacement error change rate e2 respectively; μ N (e1), μ Z (e1), μ P (e1) are the membership degrees of the negative, zero, and positive fuzzy subsets of the displacement error e1; μ N (e2), μ Z (e2), μ P (e2) are the membership degrees of the negative, zero and positive fuzzy subsets of the displacement error change rate e2 respectively.

[0133] It can be understood that fuzzification is the process of converting precise numerical inputs into fuzzy linguistic variables. In this step, the fuzzy representation of the input variables is achieved by calculating the membership degrees of the displacement error and the displacement error change rate and combining them into fuzzy variables. The vector concatenation operation is used to integrate the membership information of the displacement error and the displacement error change rate into a unified fuzzy variable, facilitating subsequent fuzzy reasoning. By converting precise numerical inputs into fuzzy linguistic variables, the control system can more flexibly handle nonlinearities and uncertainties. This flexibility helps improve the adaptability and robustness of the control system. The fuzzy variable X(t)' generated by the fuzzification operation serves as the input data for fuzzy reasoning. Through fuzzy reasoning, the system can derive the corresponding control output based on the fuzzy variables and the fuzzy rule base, achieving intelligent control of the hydraulic clamping mechanism. This mechanism provides the foundation for the automated data-driven process of the control system. Furthermore, the fuzzification operation converts precise numerical inputs into fuzzy linguistic variables, making the control system's decision-making process more intuitive and explainable. This helps technicians understand and debug the control system, improving its maintainability and reliability.

[0134] In this embodiment, if Figure 2 As shown, regarding step S2, the fuzzy controller performs fuzzy reasoning: the fuzzy controller performs fuzzy reasoning based on the fuzzy variable X(t)' at the current time step t and the rule base R, deduces possible values ​​of the output domain, and obtains the fuzzy output Fo as the control variable required for the opening of the hydraulic valve;

[0135] Specifically, step S200, rule evaluation (matching): Step S2 is the core operation stage of the fuzzy controller. It involves performing fuzzy reasoning based on the fuzzy variable X(t)' at the current time step t and the rule base R to derive possible values ​​of the output universe and derive the fuzzy output Fo, which serves as the control variable required for the hydraulic valve opening. This step uses the fuzzy reasoning mechanism to map fuzzy variables to specific control outputs, achieving intelligent control of the hydraulic clamping mechanism.

[0136] For each rule r in the rule base R i , execute the following "IF-THEN" conditional statement (used to describe the mapping relationship between the input fuzzy set and the output fuzzy set) rule matching:

[0137] "if yes and yes Then Y is ”;

[0138] Among them, the fuzzy variable X(t)' is composed of the membership vector μ of the displacement error e1 e1and the membership vector μ of the displacement error change rate e2 e2 Composition, that is, X′=[μ e1 ,μ e2 ]. In the two-dimensional fuzzy controller, X1' and X2' represent μ e1 and μ e2 , which together describe the current state of the system. X1' and X2' are components of the fuzzy variable X'(t) (two-dimensional fuzzy controller), which is the input fuzzy set, B i is the output fuzzy set, corresponding to the possible value range of the control quantity; and are input fuzzy sets, corresponding to the possible value ranges of X1' and X2' respectively;

[0139] The retrieval rule is to perform the following rule matching process for each rule ri in the rule base R:

[0140] Check whether X1' belongs to the input fuzzy set Whether it belongs to the input fuzzy set ;

[0141] If X1' and X2' meet the conditions at the same time, then the i-th rule r i Activated and ready to perform subsequent operations;

[0142] The purpose of rule matching is to find the rule that best matches the current system state for subsequent fuzzy reasoning.

[0143] It can be understood that fuzzy reasoning is a reasoning method based on fuzzy logic and fuzzy set theory. It maps input fuzzy variables to output fuzzy variables through "if-then" conditional statements in the rule base. Rule matching is one of the key steps in fuzzy reasoning. It determines which rules should be activated by checking whether the input fuzzy variables belong to the input fuzzy set in the rule. The output fuzzy set is the result of rule matching. It represents the possible range of values ​​of the control variable and provides the basis for subsequent control operations. Through fuzzy reasoning and rule matching, the system can more accurately deduce the range of values ​​of the control variable based on the current state, thereby improving control accuracy. Fuzzy reasoning can handle nonlinear and uncertain factors, allowing the system to remain stable and reliable in the face of complex and changing working environments.

[0144] Specifically, in step S201, the matching degree (activation degree) μ ​​of each rule is calculated. riIn fuzzy control systems, the matching degree (or activation degree) of a rule is used to measure the degree of compliance between the input variables and the rule conditions. For the servo-hydraulic clamping control system in this technical solution, each rule defines fuzzy set conditions for the input variables X1′ (displacement error e1) and X2′ (displacement error change rate e2).

[0145] For the i-th rule, its matching degree μ ri Defined as input variables X1′ and X2′ belonging to the fuzzy sets defined in the rules and The minimum value of the membership degree. That is: ;

[0146] in, and X1' and X2' belong to the input fuzzy set respectively and The degree of membership.

[0147] It can be understood that by calculating the matching degree of each rule, the system can more accurately assess the degree of conformance between input variables and rule conditions, thus providing an accurate basis for subsequent output aggregation. Using membership functions to describe the membership relationship between input variables and fuzzy sets enables the system to handle nonlinear and uncertain factors, improving the flexibility and robustness of control.

[0148] Specifically, step S202, aggregation: for each output fuzzy set B i , calculate its aggregated membership function ; This is achieved by taking the maximum value of the product of the matching degree and the corresponding output membership degree in all rules. That is: ;

[0149] in, The output value y belongs to the fuzzy set B i The membership degree of y i is the possible value of the fuzzy output belonging to the maximum possible case;

[0150] Further, defuzzification: convert the fuzzy output Y' into a control quantity ;

[0151] Among them, F o It is the final control quantity, which is used to adjust the opening of the hydraulic valve.

[0152] It is understandable that through aggregate calculation, the system can comprehensively consider the influence of all rules and make intelligent decisions based on the actual situation of the input variables. Using the center of gravity method to convert the fuzzy output into a specific control quantity can ensure the smooth change of the control quantity and improve the stability of the system. Since the system can dynamically adjust the control quantity according to the changes in the input variables, it has strong adaptability and can handle nonlinear and uncertain factors. Steps S201 and S202 together constitute the core part of the fuzzy backward time domain control method. By calculating the matching degree of the rules and the aggregate output, precise control of the hydraulic loosening mechanism is achieved. This method not only improves the accuracy and stability of the control, but also enhances the flexibility and adaptability of the system, and provides strong support for the automated data-driven process of the servo hydraulic clamping control system.

[0153] In this embodiment, regarding step S3, the fuzzy controller performs electrical signal conversion: the main task of this step is to convert the fuzzy output F o The defuzzification is performed and then converted into a control signal e(t), which is then transmitted to the PLC controller to adjust the opening and closing time ratio or frequency of the hydraulic valve.

[0154] Fuzzy output F o It is the result of the fuzzy controller's reasoning based on the input variables and fuzzy rules. It is usually a fuzzy set that represents the possible range of control variables and their membership. The fuzzy output Fo is defuzzified using the centroid method and then converted into a control signal e(t). The control signal e(t) is then transmitted to the PLC controller: ;

[0155] The control signal e(t) is an electrical signal that adjusts the opening and closing time ratio or frequency of the hydraulic valve. Integration is performed over the entire domain of the fuzzy output variable. By adjusting the opening and closing time ratio or frequency of the hydraulic valve, the flow and pressure of the hydraulic system can be controlled, thereby achieving precise control of the servo-hydraulic clamping mechanism.

[0156] It can be understood that defuzzification can convert fuzzy outputs into deterministic control signals, improving control precision and accuracy. Defuzzification considers all possible values ​​in a fuzzy set and their membership, and calculates the center of gravity to obtain the most representative control variable, thereby improving control precision. Fuzzy controllers can handle nonlinearity and uncertainty, and through fuzzy reasoning and defuzzification, they generate stable control signals, enhancing system stability. Fuzzy controllers handle the uncertainty and nonlinearity of input variables through fuzzy rule reasoning, and defuzzification converts fuzzy quantities into deterministic quantities, resulting in a stable control signal.

[0157] Furthermore, the combination of a fuzzy controller and a PLC controller enables an automated, data-driven process from fuzzy reasoning to control signal transmission, improving the system's automation level. The fuzzy controller performs fuzzy reasoning and defuzzification based on the input variables to generate the control signal; the PLC receives the control signal and adjusts the hydraulic valve's operating state, achieving an automated control process from input to output.

[0158] In this embodiment, regarding step S4, timing control: the main task of this step is for the PLC controller to control the opening and closing timing of the hydraulic valve according to the received control signal e(t), thereby achieving precise control of the hydraulic cylinder and completing the clamping or loosening of the workpiece.

[0159] Based on the preset complete opening and closing cycle T of the hydraulic valve, the hydraulic valve remains open for a period of time T within one cycle. on And the closing time T within a cycle off Calculated as: ;

[0160] Where, the control signal e(t) varies between 0 and 1, indicating the opening ratio. When e(t) = 1, the hydraulic valve is fully open; when e(t) = 0, the hydraulic valve is fully closed; and when e(t) is between 0 and 1, the hydraulic valve is partially open.

[0161] Among them, by controlling the opening and closing timing of the hydraulic valve, the extension and retraction of the hydraulic cylinder can be accurately controlled to achieve the clamping or loosening of the workpiece. When the workpiece needs to be clamped, the PLC controller increases the value of the control signal e(t) and extends the opening time T of the hydraulic valve. on , so that the hydraulic cylinder extends and clamps the workpiece. When the workpiece needs to be released, the PLC controller reduces the value of the control signal e(t) and shortens the opening time T of the hydraulic valve. on , causing the hydraulic cylinder to retract and release the workpiece.

[0162] As you can see, by controlling the opening and closing timing of the hydraulic valve, the extension and retraction of the hydraulic cylinder can be precisely controlled, thereby improving the accuracy of workpiece clamping or loosening. Because the control signal e(t) varies between 0 and 1, the opening ratio of the hydraulic valve can be finely adjusted, thus achieving precise control of the hydraulic cylinder.

[0163] Furthermore, the timing control method can flexibly adjust the opening and closing timing of the hydraulic valve according to different workpieces and control requirements, meeting various control needs. This is because by changing the value of the control signal e(t), the opening and closing time of the hydraulic valve can be easily adjusted, thus achieving flexible control of the system.

[0164] Furthermore, the sequential control method directly links the hydraulic valve's opening and closing timing to the control signal e(t), simplifying the control process and improving efficiency. This is because, using pre-set formulas for calculating opening and closing times, the PLC controller can directly calculate the hydraulic valve's opening and closing timing based on the control signal e(t), eliminating the need for complex calculations or judgments.

[0165] Furthermore, the timing control method combined with the fuzzy controller can effectively handle nonlinear and uncertain factors in the hydraulic system, such as fluctuations in hydraulic oil temperature and pressure. This is because the fuzzy controller can adjust the value of the control signal e(t) in real time according to changes in the input variables, thereby adapting to the nonlinear and uncertain changes in the hydraulic system.

[0166] In this embodiment, regarding step S5, a backoff time domain (mechanism) is executed: that is, a process of using past and current information to predict and adjust future control signals to ensure that the state of the hydraulic cylinder is consistent with the desired state: the sensor monitors the state of the hydraulic cylinder in real time and transmits the feedback signal f(t) consisting of the new piston position l' and the new pressure p' to the fuzzy controller; the fuzzy controller evaluates the control effect based on the feedback signal f(t); if there is an error, the backoff time domain is executed, and the fuzzy controller adjusts the control strategy to correct the error in the next time step t+1.

[0167] S500, assume that the desired piston position corresponding to the opening of the hydraulic valve controlled in step S4 is l d and the expected pressure is p d ; then calculate the position error e l (t) and pressure error e p (t):

[0168] e l (t) = l'-l d ;

[0169] e p (t) = p'-p d ;

[0170] S501, define the error vector E(t) = [e l (t), e p (t)]; where e l (t) and e p (t) are the errors of position and pressure, respectively;

[0171] S502, the fuzzy controller performs the following operations according to the error vector E(t):

[0172] S5020 Fuzzification: Calculate the membership degree of the error E(t) according to the format of steps S100-S101 and convert it into a fuzzy set. This step is to convert the error information into a fuzzy quantity that the fuzzy controller can handle.

[0173] S5021 Rule Evaluation: Evaluate the fuzzy output according to the fuzzy logic rule base R, following the format of steps S200 to S204. This step is to reason on the fuzzy set according to the rules in the fuzzy rule base and obtain the fuzzy output.

[0174] S5022 Defuzzification: According to the form of step S3, the fuzzy output is converted into a new control signal e(t+1). This step converts the fuzzy output into a deterministic electrical signal for control in the next time step.

[0175] The updated control signal e(t+1) is used in the next time step t+1 to correct the error and adjust the opening and closing timing of the hydraulic valve.

[0176] It can be understood that through real-time monitoring and feedback adjustment, the back-off time-domain mechanism can promptly detect and correct control errors, thereby improving the control accuracy of the hydraulic cylinder. The error vector E(t) contains position and pressure error information. The fuzzy controller adjusts the control strategy based on this information, bringing the actual state of the hydraulic cylinder closer to the desired state. The back-off time-domain mechanism can handle nonlinearities and uncertainties, enhancing system stability by adjusting the control strategy in real time. The fuzzy controller has strong nonlinear processing and adaptive capabilities, and can dynamically adjust the control strategy based on error information, thereby stabilizing the system state. The back-off time-domain mechanism enables autonomous detection and adjustment of the hydraulic release mechanism without manual intervention. Sensors monitor the hydraulic cylinder state in real time and transmit feedback signals to the fuzzy controller, which automatically adjusts the control strategy based on the feedback signals, forming a closed-loop control system.

[0177] In this embodiment, if Figure 5 As shown, step S6 is executed cyclically: steps S1 to S5 are executed cyclically, enabling autonomous detection and adjustment of the hydraulic release mechanism until the workpiece is machined and released. This ensures that the hydraulic release mechanism can continuously and accurately respond to control commands, handle nonlinear and uncertain factors, and maintain a stable clamping state for the workpiece during machining.

[0178] The loop execution of steps S1 to S5 is as follows:

[0179] Step S1: Fuzzify the input variables and convert the physical quantities such as the actual piston position and pressure monitored by the sensor into fuzzy quantities to provide a basis for subsequent fuzzy reasoning.

[0180] Step S2: Evaluate the rules according to the fuzzy logic rule base and obtain the fuzzy output through fuzzy reasoning. This step is the core of the fuzzy controller and is responsible for handling nonlinear and uncertain factors.

[0181] Step S3: Defuzzify the fuzzy output and convert it into a certain electrical signal as the basis for controlling the opening and closing of the hydraulic valve.

[0182] Step S4: Timing control, the PLC controller controls the opening and closing timing of the hydraulic valve according to the received control signal to achieve precise control of the hydraulic cylinder. Step S5: Execute the backward time domain mechanism, use past and current information to predict and adjust future control signals, monitor the status of the hydraulic cylinder in real time through sensors, and adjust the control strategy according to the feedback signal to correct the error. At the same time, the distribution diagram of the dependent variable under different independent variables can be visualized based on the passage of time, such as Figure 4 As shown, help staff make decisions.

[0183] The cycle continues until the workpiece is machined and released. A signal indicating completion can be provided by a sensor or an external control system; upon receipt, the cycle terminates. Throughout the cycle, the hydraulic release mechanism monitors its status in real time via sensors and autonomously adjusts its control strategy using a fuzzy controller. This autonomous detection and adjustment capability enables the hydraulic release mechanism to adapt to various changes during the machining process and maintain a stable clamping state.

[0184] It can be understood that by looping through steps S1-S5, the hydraulic clamping mechanism is able to continuously and accurately respond to control commands, improving control accuracy and stability. The fuzzy controller has powerful nonlinear processing and adaptive capabilities, and can dynamically adjust the control strategy based on the actual state, thereby maintaining precise control of the hydraulic cylinder. The autonomous detection and autonomous adjustment capabilities during the loop process enable the hydraulic clamping mechanism to handle nonlinear and uncertain factors, enhancing the robustness of the system. The fuzzy controller uses a fuzzy logic rule base for reasoning, capable of handling complex input situations and deriving reasonable control outputs, thereby adapting to various processing environments. The automated loop process reduces manual intervention and improves production efficiency.

[0185] Example 2: Based on step S200 of the embodiment, this embodiment further provides a rule base R:

[0186] Among them, the input variable X1′ is the displacement error e1, and the input variable X2′ is the displacement error change rate e2. The output variable Y is the opening control value F of the hydraulic valve o ; The rule base R is as follows:

[0187] "if yes and yes Then Y is ”

[0188] This rule states that when the displacement error is negative (indicating that the actual position is below the target position) and the rate of change of the displacement error is also negative (indicating that the error is increasing, meaning the actual position is moving away from the target position), the control variable is set to a small value (S). This is done to slow down the adjustment of the hydraulic valve opening to avoid overreaction, as rapid adjustments could cause system instability due to the increasing error.

[0189] "if is N and If it is Z, then Y is M”;

[0190] This rule states that when the displacement error is negative but the displacement error change rate is zero (indicating that the error remains unchanged), the control variable is set to medium (M). This is to moderately adjust the hydraulic valve opening to gradually reduce the error.

[0191] "if is N and If it is P, then Y is L”;

[0192] This rule states that when the displacement error is negative but the displacement error rate of change is positive (indicating that the error is decreasing, meaning the actual position is approaching the target position), the control amount is set to large (L). This is to speed up the adjustment process and eliminate the error as quickly as possible.

[0193] "if is Z and If it is N, then Y is M”;

[0194] This rule states that when the displacement error is zero but the rate of change of the displacement error is negative (indicating that the system is moving away from the target position), the control variable is set to medium (M). This is to preventively adjust the hydraulic valve opening to prevent the error from increasing.

[0195] "if is Z and If it is Z, then Y is M”;

[0196] This rule states that when the displacement error and the displacement error change rate are both zero, the control variable remains at the middle (M). This is to maintain system stability and avoid unnecessary adjustments.

[0197] "if is Z and If it is P, then Y is L”;

[0198] This rule states that when the displacement error is zero but the displacement error change rate is positive (indicating that the system is approaching the target position but too quickly), the control variable is set to large (L). This is to moderately slow down the adjustment speed and prevent overshoot.

[0199] "if is P and If it is N, then Y is L”;

[0200] This rule states that when the displacement error is positive (indicating that the actual position is higher than the target position) and the displacement error rate of change is negative (indicating that the error is increasing, meaning the actual position is moving away from the target position but in the opposite direction), the control variable is set to large (L). This is to quickly reduce the error and return the system to the target position.

[0201] "if is P and If it is Z, then Y is L”;

[0202] This rule states that when the displacement error is positive but the rate of change of the displacement error is zero, the control variable is set to large (L). This is to continuously reduce the error and return the system to the target position as quickly as possible.

[0203] "if is P and If it is P, then Y is L”;

[0204] This rule states that when the displacement error is positive and the rate of change of the displacement error is also positive (indicating that the error is increasing and the actual position is moving away from the target position), the control variable is set to very large (VL). This is to quickly and significantly adjust the hydraulic valve opening to eliminate the error as quickly as possible and prevent system instability.

[0205] It's important to note that by meticulously partitioning the input variables into fuzzy sets and assigning specific control outputs based on these combinations, the system can more accurately respond to changes in displacement error and error rate of change. The rule base considers all possible combinations of error and error rate of change and provides corresponding control strategies, helping the system maintain stability in complex and changing operating environments. The rule base simulates the decision-making process of human experts, enabling the system to automatically adjust control strategies based on real-time data, achieving intelligent control.

[0206] Example 3: Based on steps S201 to S202 of Example 1, this example further provides a push notification method for discrete situations:

[0207] In step S202, the defuzzification uses the center of gravity method (or weighted average method); however, this method is difficult to apply to the scenario applicable to this solution, that is, when the control axis cannot perform position control (servo off state), the CNC machine tool will reflect the mechanical movement error count to the current position, which will cause the error accumulation when moving relative to the instruction; and this accumulation is discrete, and the center of gravity method alone cannot handle such discrete data. The discrete process further increases the error of each output value yi Therefore, it is necessary to carry out push-in-the-heart drive and process this discretized uncertainty data. Therefore, the solution of this embodiment is:

[0208] Specifically, step S203, execute push: use DS evidence theory algorithm to output the best possible value y in the domain i ':

[0209] S2030, define basic probability assignment: for each possible output value y i , define the basic probability assignment m(y i ),satisfy and ; where Θ is the set of all possible values ​​of the output domain, which can be drawn up based on historical data.

[0210] S2031, combine evidence using Dempster's combination principle: for the i-th and j-th fuzzy output possible values ​​y i and y j , take the two as evidence sources, namely m1 and m2, then the basic probability of the combination is assigned to m 1,2 (y i ) for: ;

[0211] Where K is a normalization constant, defined as: ;

[0212] S2032, calculate the trust function Bel(y i ), indicating the output value y i The degree of belief that the belief is true: ;

[0213] S2033, calculate the likelihood function Pl(y i ), indicating the output value y i The degree of confidence that the trust is not false: ;

[0214] S2034, decision and assignment: Based on the trust function and likelihood function, use the combination of trust and uncertainty to calculate each possible output value y i Overall rating:

[0215] Select the y with the highest score i (Even if Bel(y i ) maximum) as the best possible value y i ':

[0216] ;

[0217] It is understandable that the centroid method as described in Example 1 is a commonly used fuzzy number processing method, which represents the center, position and size of a fuzzy set by calculating the centroid position of the fuzzy number. However, the centroid method is mainly suitable for processing continuously changing fuzzy data. In the discrete case, the centroid method is difficult to apply directly because it requires a continuous membership function to calculate the centroid. The error accumulation of the CNC machine tool in the servo off state is discrete, and each output value y i The membership function may not be directly described by a continuous mathematical expression, so the centroid method cannot directly process this discrete data (to be precise, it cannot process this discrete data continuously). However, when faced with multiple sources of evidence in this situation, the Dempster's combination principle of the DS evidence theory can combine them to obtain the basic probability assignment after the combination. The combination formula takes into account the conflicts and correlations between the evidence sources, and ensures that the basic probability assignment after the combination still meets the normalization conditions through the normalization constant K. Each output value y can be evaluated through the trust function and the likelihood function. i The uncertainty of , and select the output value with the highest confidence as the best possible value. Figure 3 The MATLAB integrated fuzzy control module shown in the figure tests the center of gravity method used in Example 1 (parts C and D in the figure) and the DS evidence theory used in this embodiment (parts A and B in the figure). It can be clearly seen that as time goes by (A→B, and C→D), under the same horizontal axis control amount, the vertical axis deviation brought by the DS evidence theory is significantly smaller than that in Example 1.

[0218] It is further worth pointing out that DS evidence theory has significant advantages in dealing with uncertainty. It can not only distinguish between "uncertainty" and "unknown", but also quantify uncertainty through the belief function and likelihood function:

[0219] (1) Distinguishing between "uncertainty" and "not knowing": In DS evidence theory, "uncertainty" indicates a low degree of confidence in a proposition, but not a complete denial; "not knowing" indicates a lack of information or evidence for a proposition. The above-mentioned confidence function and likelihood function can be used to quantify the uncertainty of each proposition. The confidence function represents the degree of confidence that the proposition is true, and the likelihood function represents the degree of confidence that the proposition is not false. These two functions together constitute an uncertainty interval.

[0220] (2) Integrating information from multiple sources of evidence: DS evidence theory can integrate information from multiple sources of evidence, combine them through Dempster's combination principle, and obtain a comprehensive degree of confidence in each proposition. This method has higher flexibility and robustness in dealing with uncertainty problems. Therefore, DS evidence theory effectively handles the discrete error accumulation and uncertainty problems of CNC machine tools in the servo-off state by defining basic probability assignments, using Dempster's combination principle, and calculating confidence functions and likelihood functions. Compared with the center of gravity method, DS evidence theory is more suitable for processing such discrete and uncertain data, providing a more reliable and accurate decision-making basis for the servo hydraulic clamping control method.

[0221] Specifically, step S204, defuzzification: ;

[0222] Among them, Fo is the fuzzy output, marking is its membership function, and x is the possible value range of the fuzzy output variable.

[0223] It can be understood that through the defuzzification operation, the system can convert the fuzzy output into a specific control quantity and achieve precise control. Since the defuzzification operation can quickly calculate the control quantity, it can improve the control response speed of the system. The defuzzification operation can handle fuzzy outputs in different discrete situations and enhance the adaptability of the system. Steps S203 and S204 together constitute the key part of the fuzzy backoff time domain control method. By executing the push signal and defuzzification operations, the system can pre-determine whether error accumulation occurs, handle nonlinear and uncertainty factors, and automatically modify the electrical signal and PLC timing to realize an automated data-driven process of autonomous detection and autonomous adjustment. This method not only improves the accuracy and stability of control, but also enhances the flexibility and adaptability of the system.

[0224] Experimental Example: This example aims to compare the control effects of a servo-hydraulic clamping control method based on fuzzy backoff time-domain control (experimental group) and a traditional solution (control group) in the hydraulic clamping system of a CNC machine tool. Finite element simulation is used to analyze the error accumulation rate and maximum axial offset to verify the superiority of the fuzzy backoff time-domain control method.

[0225] (1) Experimental equipment and materials:

[0226] CNC programs (G-code) edited by Mazak;

[0227] Autodesk Inventor software finite element analysis module;

[0228] CK6130 CNC machine tool modeling (only the hydraulic clamping system is retained);

[0229] (2) Experimental methods:

[0230] 2.1 Experimental Group:

[0231] A servo-hydraulic clamping control method based on fuzzy back-off time-domain control was employed. The CNC program (G-code) edited in Mazak was imported into ANSYS finite element software to develop the clamping path. A model of the CK6130 CNC machine tool was imported into the software, retaining only the hydraulic clamping system. A servo control system shutdown state was developed in the G-code to simulate the servo shutdown state of a real CNC machine tool.

[0232] Compensation control is implemented according to the logic of the fuzzy backoff time domain control method as described in embodiments one to three.

[0233] 2.2 Control Group:

[0234] A traditional solution was adopted. A servo-hydraulic clamping control method based on fuzzy back-off time-domain control was employed. The CNC program (G-code) edited in Mazak was imported into ANSYS finite element software to develop the clamping path. A model of the CK6130 CNC machine tool was imported into the software, retaining only the hydraulic clamping system. A servo control system shutdown state was developed in the G-code to simulate the servo shutdown state of a real CNC machine tool.

[0235] After receiving the clamping signal and detecting the axis in position signal, the hydraulic valve is opened after a delay of 500ms; after detecting the pressure relay signal, the axis prohibition signal is given after a delay of 200ms to turn off the axis enable.

[0236] When the release command is detected, the axis prohibition is canceled first, the axis enable is loaded, and the valve is closed after a delay of 200ms; when the pressure relay signal is detected to be 0, the axis rotation command is loaded after a delay of 500ms.

[0237] Compensation control is implemented in ANSYS software according to the logic of traditional methods.

[0238] (3) Simulation and analysis:

[0239] A 330-minute time interval was simulated in the software. The error accumulation rate and maximum axial deviation of the experimental and control groups were analyzed based on finite element analysis.

[0240] (IV) Experimental results:

[0241] 4.1 Error accumulation rate:

[0242] like Figure 6 As shown in Figure 2, the deviation rate of the experimental group is significantly smaller than that of the control group. This indicates that within the same time period, the experimental group's method can more effectively control the accumulation of errors.

[0243] 4.2 Maximum axial offset:

[0244] like Figure 7-Figure 8 As shown in Figure 2, finite element simulation analysis showed that the degree of deviation in the experimental group was significantly smaller than that in the control group, which further demonstrated the superiority of the experimental group method in controlling axial deviation.

[0245] (V) Experimental analysis

[0246] 5.1 Analysis of control group problems:

[0247] In the control group, the axis experienced mechanical movement while position control was disabled. This mechanical movement caused mechanical displacement feedback pulses to accumulate in the error counter. The CNC machine tool reflected this mechanical movement error count in the current position, causing errors to accumulate relative to the commanded movement. Each time the servo was shut off, the error accumulated and gradually increased until positioning accuracy exceeded tolerance.

[0248] 5.2 Advantages of the experimental group:

[0249] The fuzzy backoff time-domain control method employed by the experimental group was able to monitor and adjust the control strategy in real time, effectively suppressing the accumulation of errors. Through fuzzy logic reasoning and adaptive adjustment capabilities, the experimental group's method was able to better adapt to various changes in the machining process and maintain a stable clamping state.

[0250] (6) Conclusion:

[0251] This study compared the control performance of a servo-hydraulic clamping control method based on fuzzy backoff time-domain control with a traditional solution in a CNC machine tool hydraulic clamping system. The experimental group achieved significantly lower error accumulation rates and greater maximum axial offset than the control group. The control group experienced mechanical movement when the control axis was unable to maintain position control, leading to continuous error accumulation. The fuzzy backoff time-domain control method employed by the experimental group demonstrated superior control effectiveness and adaptability.

[0252] Therefore, this experiment verifies the superiority of the servo hydraulic clamping control method based on fuzzy backward time domain control, and provides an effective solution for the control of the hydraulic clamping system of CNC machine tools.

[0253] All of the above embodiments merely represent implementation methods of the present invention in practical applications. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the scope of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the appended claims.

[0254] For those skilled in the art, it can be further appreciated that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0255] At the same time, those skilled in the art will understand that all or part of the processes in all the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

Claims

1. A servo hydraulic clamping control method based on fuzzy back-off time domain control includes a sensor that monitors the piston position l and pressure p of the hydraulic cylinder in real time and converts them into digital signals, characterized in that: The following steps are included: S1, the fuzzy controller receives the control axis variable X(t)=[e1,e2] transmitted by the CNC control system at the current time step t, where e1 is the displacement error and e2 is the displacement error change rate; the control axis variable X(t) is converted into a fuzzy variable X(t)' using the membership function; S2, the fuzzy controller performs fuzzy reasoning based on the fuzzy variable X(t)' at the current time step t and the rule base R, and deduces the possible values ​​of the output domain, and obtains the fuzzy output Fo as the control variable required for the opening of the hydraulic valve; the fuzzy reasoning method is: S201, for each rule r in the rule base R i , execute the following "IF-THEN" conditional statement rule matching: "If yes and yes Then Y is ” Among them, X1' and X2' are components of the fuzzy variable X'(t), which is the input fuzzy set, B i is the output fuzzy set; and is the input fuzzy set; S201, calculate the matching degree μ of each rule ri : ； in, and X1' and X2' belong to the input fuzzy set respectively and The degree of membership; S202, for each output fuzzy set B i , calculate its aggregated membership function : ； in, The output value y belongs to the fuzzy set B i The membership degree of y i is the possible value of the fuzzy output belonging to the maximum possible case; The method of inference is to use the DS evidence theory algorithm to output the best possible value y in the domain i ': S2030, for each possible output value y i , define the basic probability assignment m(y i ), satisfying m(∅)=0 and ; where Θ is the set of all possible values ​​of the output domain; S2031, for the i-th and j-th fuzzy output possible values ​​y i and y j , take the two as evidence sources, namely m1 and m2, then the basic probability of the combination is assigned to m 1,2 (y i ) for: ； Where K is a normalization constant, defined as: ; S2032, calculate the trust function Bel(y i ): ; S2033, calculate the likelihood function Pl(y i ): ; S2034, select the y with the highest score i As the best possible value y i ': ； S3, defuzzifies the fuzzy output Fo using the center of gravity method, converts it into a control signal e(t), and transmits the control signal e(t) to the PLC controller; S4, the PLC controller controls the opening and closing timing of the hydraulic valve according to the received control signal e(t); In step S5, the fuzzy controller evaluates the control effect based on the feedback signal f(t) consisting of the new piston position l' and the new pressure p'; ​​if there is an error, the backward time domain is executed, and the fuzzy controller adjusts the control strategy to correct the error in the next time step t+1.

2. The servo hydraulic clamping control method according to claim 1, characterized in that: In S1, the membership of the displacement error e1 and the displacement error change rate e2 is calculated; the fuzzy set of the displacement error e1 and the displacement error change rate e2 is divided into three fuzzy subsets: negative N, zero Z and positive P: Membership function of negative N fuzzy subsets ; Membership function of zero-Z fuzzy subsets ; Membership function of positive P fuzzy subsets ; Membership function of negative N fuzzy subsets ; Membership function of zero-Z fuzzy subsets ; Membership function of positive P fuzzy subsets ; Among them, c1, c2, c3 are the centers of negative, zero, and positive fuzzy subsets respectively, and σ1, σ2, and σ3 are their standard deviations respectively; c4, c5, c6 are the centers of negative, zero, and positive fuzzy subsets respectively, and σ4, σ5, and σ6 are their standard deviations respectively.

3. The servo hydraulic clamping control method according to claim 2, characterized in that: In S1, the method of forming the fuzzy variable X(t)' is: ; Among them, [·,·] represents the vector concatenation operation.

4. The servo hydraulic clamping control method according to claim 3, characterized in that: In S2, the fuzzy output Fo is obtained by: ; Among them, Fo is the fuzzy output, marking is its membership function, and x is the possible value range of the fuzzy output variable.

5. The servo hydraulic clamping control method according to any one of claims 1 to 4, characterized in that: In S3, the method of converting into the control signal e(t) is: ; The control signal e(t) is an electrical signal that adjusts the opening and closing time ratio or frequency of the hydraulic valve.

6. The servo hydraulic clamping control method according to claim 5, characterized in that: The execution steps of S5 include: S500, assume that the desired piston position corresponding to the opening of the hydraulic valve controlled in step S4 is l d and the expected pressure is p d ; then calculate the position error e l (t) and pressure error e p (t): e l (t) = l'-l d ; e p (t) = p'-p d ; S501, define the error vector E(t) = [e l (t), e p (t)]; where e l (t) and e p (t) are the errors of position and pressure, respectively; S502, the fuzzy controller performs the following operations according to the error vector E(t): S5020, calculating the membership degree of the error E(t) and converting it into a fuzzy set; S5021, Evaluate fuzzy output; S5022, defuzzification: According to step S4, the fuzzy output is converted into a new control signal e(t+1); the control signal e(t+1) is updated for the next time step t+1 to correct the error.

7. The servo hydraulic clamping control method according to claim 6, characterized in that: The process also includes S6, which is a cyclic execution: cyclic execution of steps S1 to S5 to achieve autonomous detection and autonomous adjustment of the hydraulic clamping mechanism until the workpiece is processed and released and removed.

8. A servo hydraulic clamping control system based on fuzzy back-off time domain control, characterized in that: include: A fuzzy controller for receiving the control axis variable X transmitted by the CNC control system and performing fuzzification operations, fuzzy reasoning and electrical signal conversion; A PLC controller for receiving the control signal e(t) calculated by the fuzzy controller and controlling the opening and closing timing of the hydraulic valve; A sensor unit mounted on the piston rod, used to feed back the detected control axis variable X to the CNC control system and to feed back the feedback signal f(t) to the fuzzy controller; A memory storing program instructions and a controller connected to the memory, when the controller executes the program instructions, causes the fuzzy controller to execute the servo hydraulic clamping control method according to any one of claims 1 to 7.

Images