Control method for servo system

Abstract

The invention relates to the technical field of follow-up system position control, in particular to a control method for a follow-up system, which comprises the following steps of: performing first self-adaptive dynamic correction on a proportionality coefficient of a PID (Proportion Integration Differentiation) controller in the follow-up system through a small error section hyperbolic cosine formula, constructing a fuzzy controller, and before the follow-up system outputs a speed given control signal, performing second self-adaptive dynamic correction on the proportionality coefficient of the PID controller in the follow-up system; and according to the position error and the position error change rate which are acquired in real time, performing second-time adaptive dynamic correction on the proportionality coefficient after the first-time adaptive dynamic correction, and according to the proportionality coefficient after the second-time adaptive dynamic correction and the position error of the servo system, outputting a speed given control signal of the servo system. The method is used for real-time control of the servo system. According to the method, hyperbolic cosine adjustment and fuzzy logic control are combined, so that the follow-up system can comprehensively meet performance index requirements in the aspects of dynamic response, steady-state precision, disturbance rejection and the like, and high-precision, quick-response, stable and reliable position control is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of position control of a servo system, and in particular to a control method for a servo system. Background Art

[0002] Position control for a servo system sets the position value as the target variable, and its speed is determined by the difference between the target position and the actual position. The performance indicators of a servo system cover both dynamic and static dimensions. For position control of a servo system, stability, control accuracy, dynamic response, and disturbance immunity are specific criteria for evaluating the system.

[0003] In practical engineering applications, the position control of servo systems typically uses a traditional control strategy called a compound PID with a feedforward coefficient. However, this approach often struggles to meet servo performance requirements when dealing with small error ranges in nonlinear systems.

[0004] When adjusting small error segments in the position control of a servo system, traditional PID controllers, when adjusting the proportional parameter to meet the servo system's dynamic response time requirements, may overreact if the proportional parameter is increased. This can cause the system's output target speed to frequently overshoot and drop, leading to system oscillations. Under the action of the integral link, small errors will accumulate rapidly, potentially leading to integral saturation, causing the controller's output to reach its upper limit, making it impossible to effectively regulate the system and requiring a long time for the system to return to a stable state. Although differential regulation can accelerate the system's stabilization, especially when responding to rapid changes in the target position, and can effectively suppress oscillations caused by system dynamics, it is extremely sensitive to noise and the adjustment process is relatively complex. Even tiny measurement noise can cause the target speed signal to fluctuate dramatically, triggering system oscillations. Summary of the Invention

[0005] The purpose of the present invention is to address the corresponding deficiencies of the existing technology and provide a control method for a servo system. The method uses the error between the target position and the actual position and its rate of change as the basis for control input, constructs an adaptive adjustment mechanism for the proportional coefficient based on the hyperbolic cosine function, and dynamically adjusts the proportional control intensity according to the error size through the nonlinear characteristics of the hyperbolic cosine function to achieve balanced optimization of the system response speed and stability. At the same time, the fuzzy logic control strategy is integrated, and the fuzzy rule table is used to make intelligent decisions and adjustments to the system dynamic process, effectively suppressing system oscillations and enhancing the system's anti-interference ability. The composite control method of the present invention can, through the synergistic effect of hyperbolic cosine regulation and fuzzy logic control, enable the servo system to fully meet the performance index requirements in terms of dynamic response, steady-state accuracy and anti-disturbance, thereby achieving high-precision, fast-response, stable and reliable position control.

[0006] The purpose of the present invention is to adopt the following scheme to achieve:

[0007] A control method for a servo system uses a small error segment hyperbolic cosine formula to perform a first adaptive dynamic correction on the proportional coefficient of a PID controller in the servo system, and outputs a speed reference control signal of the servo system based on the proportional coefficient after the first adaptive dynamic correction and the position error of the servo system for real-time control of the servo system.

[0008] Preferably, a fuzzy controller is constructed, and before the servo system outputs a speed given control signal, the fuzzy controller is used to perform a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction based on the position error and the position error change rate obtained in real time. Based on the proportional coefficient after the second adaptive dynamic correction and the position error of the servo system, a speed given control signal of the servo system is output for real-time control of the servo system.

[0009] Preferably, the small error segment hyperbolic cosine formula is used to perform a first adaptive dynamic correction on the proportional coefficient of the PID controller in the servo system, comprising the following steps:

[0010] ① The hyperbolic cosine formula for the small error segment is established as follows:

[0011]

[0012] Where K cosh is the proportional coefficient after the first adaptive dynamic correction; a is the factor that determines K cosh The minimum positive integer parameter; b and a together determine K cosh The maximum value is a positive integer parameter; c is a positive integer parameter that determines the rate of change of the hyperbolic cosine curve; e is the position error; and m is a correction coefficient used to adjust the positive integer parameter c.

[0013] ② According to the requirements of the servo system to be adjusted into position, determine the value range of the proportional coefficient;

[0014] ③ Adjust parameter b so that the servo system can achieve the fastest response speed when the positioning requirement is met;

[0015] ④ According to the oscillation of the servo system, adjust the parameter c to improve the stability of the small error segment and output the proportional coefficient of the first adaptive dynamic correction.

[0016] Preferably, before the servo system outputs the speed given control signal, the step of performing a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction according to the position error and the position error change rate obtained in real time includes:

[0017] 1) Obtain position error and position error change rate in real time as input variables of the fuzzy controller;

[0018] 2) Perform fuzzification processing on each input variable to obtain the fuzzy linguistic variables of each input variable;

[0019] 3) Construct a fuzzy rule table and use it to infer the fuzzy language variables of the input variables to obtain the fuzzy language of the output variables;

[0020] 4) Defuzzify the fuzzy language of the output variable to obtain the final fuzzy logic control signal;

[0021] 5) According to the final fuzzy logic control signal, the proportional coefficient of the second adaptive dynamic correction is output.

[0022] Preferably, in step 2), the specific steps of performing fuzzification processing on the input variable to obtain the fuzzy linguistic variable of the input variable include:

[0023] 2-1) Determine the domain of input variables based on the actual needs of the system;

[0024] 2-2) Divide the domain of input variables into several fuzzy sets and define corresponding fuzzy linguistic variables;

[0025] 2-3) Define the membership function of the input variables, and obtain the membership of each input variable on each fuzzy linguistic variable based on the membership function to form the fuzzy linguistic variable of the input variable.

[0026] Preferably, in step 4), the specific steps of defuzzifying the fuzzy language of the output variable using the center of gravity method to obtain the final fuzzy logic control signal include:

[0027] 4-1) Define the central value of each fuzzy linguistic variable of the output variable;

[0028] 4-2) Multiply each membership degree of the output variable by the corresponding central value and sum them to obtain a weighted sum;

[0029] 4-3) Add up all the memberships of the output variables to get the membership sum;

[0030] 4-4) According to the weighted sum and the membership sum, the final fuzzy logic control signal is obtained.

[0031] Preferably, in step 5), the expression for outputting the proportional coefficient of the second adaptive dynamic correction is as follows:

[0032] K=K cosh +K fuzzy

[0033] Where, K is the proportional coefficient after the second adaptive dynamic correction; K cosh is the proportional coefficient after the first adaptive dynamic correction; K fuzzy is the final fuzzy logic control signal.

[0034] Preferably, the speed setting control signal of the output servo system is:

[0035] u=K*e

[0036] Where u is the speed reference control signal; K is the proportional coefficient after the second adaptive dynamic correction; and e is the position error.

[0037] The beneficial effects of the present invention include the following:

[0038] A control method for a servo system uses a small error segment hyperbolic cosine formula to perform a first adaptive dynamic correction on the proportional coefficient of a PID controller in the servo system, and outputs a speed reference control signal of the servo system based on the proportional coefficient after the first adaptive dynamic correction and the position error of the servo system for real-time control of the servo system.

[0039] Preferably, a fuzzy controller is constructed, and before the servo system outputs a speed given control signal, the fuzzy controller is used to perform a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction based on the position error and the position error change rate obtained in real time. Based on the proportional coefficient after the second adaptive dynamic correction and the position error of the servo system, a speed given control signal of the servo system is output for real-time control of the servo system.

[0040] The present invention uses the hyperbolic cosine formula in the small error segment to perform the first dynamic correction on the proportional coefficient, effectively solving the stability and accuracy problems of the servo system in the small error segment. At the same time, a fuzzy controller is constructed and the fuzzy logic control signal is calculated by combining the real-time position error and the error change rate, thus achieving the second dynamic optimization adjustment of the proportional coefficient, significantly improving the dynamic response performance and tracking accuracy of the servo system in the small error segment.

[0041] The present invention further enhances the robustness and anti-interference capability of the system by online adjusting the proportional coefficient by combining the basic proportional coefficient with the fuzzy logic control signal; finally, a speed given control signal is generated according to the adjusted proportional coefficient and the position error, thereby realizing high-precision and high-stability control of the servo system, effectively reducing the system overshoot and improving the dynamic following performance.

[0042] Preferably, the small error segment hyperbolic cosine formula is used to perform a first adaptive dynamic correction on the proportional coefficient of the PID controller in the servo system, comprising the following steps:

[0043] ① The hyperbolic cosine formula for the small error segment is established as follows:

[0044]

[0045] Where K cosh is the proportional coefficient after the first adaptive dynamic correction; a is the factor that determines K cosh The minimum positive integer parameter; b and a together determine K cosh The maximum value is a positive integer parameter; c is a positive integer parameter that determines the rate of change of the hyperbolic cosine curve; e is the position error; and m is a correction coefficient used to adjust the positive integer parameter c.

[0046] ② According to the requirements of the servo system to be adjusted into position, determine the value range of the proportional coefficient;

[0047] ③ Adjust parameter b so that the servo system can achieve the fastest response speed when the positioning requirement is met;

[0048] ④ According to the oscillation of the servo system, adjust the parameter c to improve the stability of the small error segment and output the proportional coefficient of the first adaptive dynamic correction.

[0049] This invention achieves precise dynamic control of the initial correction of the servo system's proportional coefficient by optimizing the adjustment method of the hyperbolic cosine formula for small error ranges. By layering the static range and dynamic characteristics of the basic proportional coefficient through the key parameters a, b, and c in the formula, it is possible to ensure that the system is adjusted to the correct position while also balancing rapidity and stability in small error ranges. This approach not only ensures the system's ability to quickly respond to the target position, but also effectively avoids the overshoot and oscillation problems that are common with traditional proportional control in small error ranges, significantly improving the control accuracy and steady-state performance of the servo system.

[0050] Preferably, before the servo system outputs the speed given control signal, the step of performing a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction according to the position error and the position error change rate obtained in real time includes:

[0051] 1) Obtain position error and position error change rate in real time as input variables of the fuzzy controller;

[0052] 2) Perform fuzzification processing on each input variable to obtain the fuzzy linguistic variables of each input variable;

[0053] 3) Construct a fuzzy rule table and use it to infer the fuzzy language variables of the input variables to obtain the fuzzy language of the output variables;

[0054] 4) Defuzzify the fuzzy language of the output variable to obtain the final fuzzy logic control signal;

[0055] 5) According to the final fuzzy logic control signal, the proportional coefficient of the second adaptive dynamic correction is output.

[0056] This invention uses real-time position error and position error change rate as input variables to ensure that the fuzzy controller can dynamically perceive system state deviations and their changing trends, providing a data basis for subsequent precise control. By fuzzifying the input variables, precise numerical values ​​are converted into language-based fuzzy concepts, effectively addressing the uncertainty and nonlinear characteristics of the control process.

[0057] The present invention constructs a fuzzy rule table for reasoning, and can enable the controller to generate targeted control strategies according to different error state combinations based on rules formulated by expert experience, thereby achieving intelligent response to complex working conditions; and the present invention defuzzifies the fuzzy language of the output variables, so that the fuzzy reasoning results are converted into precise control signals, which not only retains the decision-making advantages of fuzzy logic, but also meets the actual system's demand for precise control quantities.

[0058] Preferably, in step 2), the specific steps of performing fuzzification processing on the input variable to obtain the fuzzy linguistic variable of the input variable include:

[0059] 2-1) Determine the domain of input variables based on the actual needs of the system;

[0060] 2-2) Divide the domain of input variables into several fuzzy sets and define corresponding fuzzy linguistic variables;

[0061] 2-3) Define the membership function of the input variables, and obtain the membership of each input variable on each fuzzy linguistic variable based on the membership function to form the fuzzy linguistic variable of the input variable.

[0062] The present invention determines the domain of input variables according to actual needs, which can match the adaptability of the fuzzy controller with the actual control needs. By dividing the domain of input variables into several fuzzy sets and defining corresponding fuzzy linguistic variables, it achieves an accurate description of the discretization of continuous variables and effectively handles the nonlinearity and uncertainty problems in the system.

[0063] By meticulously designing membership functions and accurately calculating the membership of each input variable with respect to each fuzzy linguistic variable, this invention produces a quantitative result that comprehensively reflects the fuzzy characteristics of the input variables. In other words, through a systematic fuzzification process, this invention effectively ensures that the fuzzy controller accurately characterizes the characteristics of the input variables, laying a solid foundation for high-precision control of servo systems.

[0064] Preferably, in step 4), the specific steps of defuzzifying the fuzzy language of the output variable using the center of gravity method to obtain the final fuzzy logic control signal include:

[0065] 4-1) Define the central value of each fuzzy linguistic variable of the output variable;

[0066] 4-2) Multiply each membership degree of the output variable by the corresponding central value and sum them to obtain a weighted sum;

[0067] 4-3) Add up all the memberships of the output variables to get the membership sum;

[0068] 4-4) According to the weighted sum and the membership sum, the final fuzzy logic control signal is obtained.

[0069] The present invention adopts the centroid method to defuzzify the fuzzy language of the output variable. On the one hand, it can effectively integrate the membership information of all fuzzy language variables, thereby avoiding information loss and ensuring that the output result can fully reflect the decision-making intention of fuzzy reasoning, so that the control signal is more in line with the actual needs of the system; on the other hand, its calculation process is relatively concise and efficient, and can be simplified to a weighted sum operation under a discrete domain, which is suitable for the fast calculation requirements of real-time control systems and ensures the timeliness of control signals.

[0070] Preferably, in step 5), the expression for outputting the proportional coefficient of the second adaptive dynamic correction is as follows:

[0071] K=K cosh +K fuzzy

[0072] Where, K is the proportional coefficient after the second adaptive dynamic correction; K cosh is the proportional coefficient after the first adaptive dynamic correction; K fuzzy is the final fuzzy logic control signal.

[0073] The present invention adjusts the proportional coefficient K after the second adaptive dynamic correction cosh And the final fuzzy logic control signal K fuzzy The online adjustment of the proportional coefficient K is achieved by combining the sum of the two. This not only solves the problems of in-place adjustment and partial overshoot when dealing with small errors in nonlinear systems or servo systems requiring nonlinear responses, but also improves the system's dynamic tracking response performance. This combination effectively improves the system's response speed and stability, reduces system overshoot, and enhances system robustness.

[0074] Preferably, the speed setting control signal of the output servo system is:

[0075] u=K*e

[0076] Where u is the speed reference control signal; K is the proportional coefficient after the second adaptive dynamic correction; and e is the position error.

[0077] The present invention uses a dynamic proportional coefficient K to perform real-time online adjustments based on the position error e. The product of the proportional coefficient K and the position error e generates a speed control signal u. This allows the servo system to accurately adjust K when in a small error range, ensuring that the speed control signal u responds appropriately to changes in the position error. This approach not only helps the system quickly reduce errors, improves its response speed, and achieves rapid tracking of the target position, but also effectively suppresses overshoot and enhances the system's stability during dynamic processes, ultimately making the entire servo system more precise and efficient in position control.

[0078] The advantages of the present invention are as follows:

[0079] This invention addresses the control contradictions that exist in servo system position control in small error ranges: an excessively large proportional parameter easily causes system overshoot, while a too small proportional parameter results in a slow time response. This approach effectively improves the problem by adjusting the hyperbolic cosine parameter. The nonlinear characteristics of the hyperbolic cosine function dynamically optimize the proportional control strength and adaptively adjust the control gain during the error convergence phase. This prevents both overshoot caused by an excessively strong proportional action and response lag caused by an insufficient proportional action, thereby improving the system's dynamic performance.

[0080] Furthermore, this invention replaces traditional differential regulation with fuzzy logic control, effectively addressing oscillations caused by measurement noise or sudden target changes during dynamic system changes. By building a rule base based on errors and their rates of change, fuzzy logic control replaces precise differential calculations with a linguistic, fuzzy decision-making mechanism, enabling smooth control of system dynamics and enhancing the system's robustness and anti-interference capabilities.

[0081] Furthermore, the present invention introduces adaptive control parameters that simultaneously optimize the system's steady-state error and rapid performance. By properly configuring these parameters, not only can high-precision control be achieved during steady-state operation, but the system's transient response speed can also be effectively improved, enabling the system to maintain good stability while rapidly tracking the target, thereby comprehensively enhancing the overall performance of the servo system's position control.

[0082] Glossary:

[0083] System repositioning: In this context, this refers to the system's accurate, rapid, and stable transition from its initial position to its target position. In servo system position control, this involves requirements for multiple indicators, including position error, repositioning time, overshoot, and oscillation. This means the system must not only ultimately reach the target position, but also do so within the specified timeframe, while ensuring that no excessive overshoot or severe oscillation occurs during the repositioning process, ensuring reliable and stable system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 Schematic diagram of a curve of the proportional coefficient after the first adaptive dynamic correction in this embodiment of the present invention;

[0085] Figure 2 Schematic diagram of another curve of the proportional coefficient after the first adaptive dynamic correction in this embodiment of the present invention;

[0086] Figure 3 Schematic diagram of error curves between conventional proportional control and hyperbolic cosine control in an embodiment of the present invention;

[0087] Figure 4 Schematic diagram of the membership function according to an embodiment of the present invention;

[0088] Figure 5 This is an error comparison diagram of the servo system in an embodiment of the present invention performing the same sinusoidal motion when the same hyperbolic cosine parameters and fuzzy logic are introduced simultaneously;

[0089] Figure 6 Flowchart of an embodiment of the present invention. DETAILED DESCRIPTION

[0090] like Figures 1 to 6 As shown, a control method for a servo system uses a small error segment hyperbolic cosine formula to perform a first adaptive dynamic correction on the proportional coefficient of a PID controller in the servo system, and constructs a fuzzy controller. Before the servo system outputs a speed given control signal, the fuzzy controller performs a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction based on the position error and the position error change rate obtained in real time. Based on the proportional coefficient after the second adaptive dynamic correction and the position error of the servo system, a speed given control signal of the servo system is output for real-time control of the servo system. The method specifically includes the following steps:

[0091] S1: Using the small error segment hyperbolic cosine formula, the proportional coefficient of the PID controller in the servo system is adaptively and dynamically corrected for the first time, including the following steps:

[0092] ① The hyperbolic cosine formula for the small error segment is established as follows:

[0093]

[0094] Where K cosh is the proportional coefficient after the first adaptive dynamic correction; a is the factor that determines K cosh The minimum positive integer parameter (i.e., the lower limit parameter); b is the parameter that determines K together with a. coshThe maximum positive integer parameter (i.e., gain parameter); c is a positive integer parameter that determines the rate of change of the hyperbolic cosine curve (i.e., curve shape parameter); e is the position error; and m is a correction coefficient used to adjust the positive integer parameter c.

[0095] In the above-mentioned small error segment hyperbolic cosine formula, the value of m is determined according to the actual working conditions (including the turret moment of inertia and the actual output power of the motor). When the actual output power of the motor is constant, the larger the turret moment of inertia, the larger the value of m; conversely, the smaller the turret moment of inertia, the smaller the value of m. In this embodiment, the optimal value of m obtained through actual working condition experiments is 0.1. That is, the small error segment hyperbolic cosine formula in this embodiment is:

[0096]

[0097] ② Based on the servo system's in-place adjustment requirements, the proportional coefficient (a + b) is adjusted to determine its range, providing clear constraints for subsequent parameter adjustments. At this point, parameter a is given a minimum value; ideally, it is 0. Therefore, under actual operating conditions, the smaller the value of parameter a, the better, provided the system's in-place adjustment requirements are met.

[0098] like Figure 1 As shown, Δs is the error between the target position and the actual position (i.e., position error e), K cosh The value range is [1,3], that is, the value of parameter a is given as 1, and parameter c is the empirical initial value, which is generally 1.

[0099] ③ Adjust parameter b so that the system can adjust the speed of the servo system while meeting the requirements of the rotation into position; that is, in this embodiment, parameter b=2 before adjustment and parameter b=1 after adjustment. At this time, the servo system achieves the fastest response speed.

[0100] ④ After determining parameters a and b, the system meets the rapidity requirements. However, if the system oscillates at this time, it is necessary to adjust parameter c to adjust the curve change rate of the hyperbolic cosine of the small error segment, thereby improving the stability and accuracy of the small error segment and outputting the proportional coefficient of the first adaptive dynamic correction.

[0101] like Figure 2 As shown, when parameters a and b are the same, the larger the parameter c is, the higher the K cosh The sharper the curve, the smaller the parameter c is, and K cosh The flatter the curve.

[0102] Applying hyperbolic cosine to proportional control (i.e. adjusting the proportional coefficient) can effectively handle certain nonlinear systems or situations where nonlinear responses are required. In this embodiment, the desired control effect is achieved by combining the hyperbolic cosine function with proportional control and adjusting its parameters and other control elements. Figure 3 As shown in FIG. 1 , in the position control of a certain servo system, under the same system, the same working conditions and the same servo inner loop parameters, the actual effect of handling the small error segment transfer process is shown. Figure 3 In the curve, the corresponding proportional coefficient K = 10, and (a + b) = 10. Figure 3 The left curve shows that when the system rotates 15mil and meets the rotation time requirement, there is oscillation; Figure 3 The curve on the right side of the middle graph is the actual effect diagram of introducing hyperbolic cosine. It can be clearly seen that by fine-tuning the proportional coefficient through the hyperbolic cosine function, it can simultaneously meet the requirements of adjusting to the position and eliminating system overshoot.

[0103] However, it is difficult to solve the system dynamic tracking response performance by hyperbolic cosine alone, so fuzzy logic control is introduced to improve the dynamic response performance of the system and thus enhance the system robustness.

[0104] S2: Construct a fuzzy controller that can dynamically adjust the proportional coefficient K online based on fuzzy theory. Before the servo system outputs a speed control signal, the fuzzy controller is used to obtain the final fuzzy logic control signal based on the position error and position error change rate obtained in real time. This allows for a second adaptive dynamic correction of the proportional coefficient after the first adaptive dynamic correction, and outputs the second adaptive dynamic correction proportional coefficient. This specifically includes the following steps:

[0105] 1) When building a fuzzy controller, its key components need to be determined. Define the input variables as the position error and the rate of change of the position error, and define the output variable as the fuzzy logic control signal K fuzzy The position error and the rate of change of the position error are obtained in real time as input variables of the fuzzy controller. The position error and the rate of change of the position error are obtained as follows:

[0106]

[0107] Where, e is the position error, that is, the difference between the target position and the current position; p ref is the position given; p loc is the current position; ec is the rate of change of position error, that is, the change of position error per unit time; e (k) is the position error at the kth moment; e (k+1) is the position error at the (k+1)th moment.

[0108] 2) In order to implement fuzzy control, fuzzy linguistic variables and domains of input and output variables must be set. In this embodiment, the specific method of fuzzifying each input variable and obtaining the fuzzy linguistic variables of each input variable includes:

[0109] 2-1) Determine a reasonable domain of input variables based on the actual system requirements (such as the system's physical limits, control accuracy, dynamic characteristics, and debugging experience) and the small error function segment. In this embodiment, the domain of the position error e is {-240, -160, -80, 0, 80, 160, 240}, and the domain of the position error change rate ec is {-60, -40, -20, 0, 20, 40, 60}.

[0110] 2-2) Divide the universe of the input variables (i.e., position error e and position error change rate ec) into seven fuzzy sets, each corresponding to a fuzzy linguistic variable. In this solution, seven universal fuzzy linguistic variables are used to form the fuzzy sets:

[0111] {NB, NM, NS, ZO, PS, PM, PB}

[0112] They represent Negative Big, Negative Medium, Negative Small, Zero, Positive Small, Positive Medium, and Positive Big respectively.

[0113] 2-3) Define the membership function of the input variable, which can express the relationship between the specific value of the input variable and the fuzzy linguistic variable. In this embodiment, the membership of each input variable on each fuzzy linguistic variable is obtained according to the membership function to form the fuzzy linguistic variable of the input variable. Figure 4 As shown in the figure, the membership functions of the input variable position error e and the position error change rate ec are respectively displayed, which can map specific values ​​to fuzzy linguistic variables and effectively realize fuzzification.

[0114] 3) A fuzzy rule table is constructed based on the small error segment characteristics of the servo system. As an important part of fuzzy logic control, the fuzzy rule table directly determines the control performance of the servo system. In this embodiment, the fuzzy logic rule table adopted is as follows:

[0115] Table 1 Fuzzy rule base

[0116]

[0117] The fuzzy rule table is used to infer the fuzzy language variables of the input variables to obtain the output variable Kfuzzy For example, the fuzzy result of the position error e after fuzzification is NB and NM, and the fuzzy result of the position error change rate ec after fuzzification is NM and NS. By looking up Table 1, we can infer that the output fuzzy linguistic variables are two fuzzy linguistic variables, namely PB and PM.

[0118] 4) Defuzzify the fuzzy language of the output variable using the center of gravity method to obtain the final fuzzy logic control signal. The specific steps include:

[0119] 4-1) Define the central value of each fuzzy linguistic variable of the output variable. In this embodiment, the domain of the output variable is {-3, -2, -1, 0, 1, 2, 3}, that is, the central value of its fuzzy linguistic variable is NB = -3, NM = -2, NS = -1, ZO = 0, PS = 1, PM = 2, PB = 3.

[0120] 4-2) Multiply each membership degree of the output variable by the corresponding central value and sum them to obtain the weighted sum, that is:

[0121] 4-3) Add up all the memberships of the output variables to get the membership sum, that is: In this embodiment, the calculated value is 1.

[0122] 4-4) According to the weighted sum and the membership sum, the final fuzzy logic control signal is obtained, which is expressed as follows:

[0123]

[0124] Where K fuzzy is the final fuzzy logic control signal (i.e., the correction coefficient of the proportional coefficient); i is the central value of the i-th output fuzzy linguistic variable; μ e is the membership degree of position error; μ ec is the membership degree of the position error change rate. It is worth noting that the above z i It is determined by the output membership function.

[0125] During the dynamic response of the servo system, when the tracking error is large, the control amount is increased quickly to reduce the error; when the tracking error is small, the control amount is slowly reduced to prevent overshoot.

[0126] 5) According to the final fuzzy logic control signal, the proportional coefficient of the second adaptive dynamic correction is output, that is, the proportional coefficient K after the first adaptive dynamic correction is calculated cosh And the final fuzzy logic control signal K fuzzy The sum (the final fuzzy logic control signal K fuzzyDirectly acting on the hyperbolic function), the proportional coefficient K is corrected quadratically, and the expression is as follows:

[0127] K=K cosh +K fuzzy

[0128] Where, K is the proportional coefficient after the second adaptive dynamic correction; K cosh is the proportional coefficient after the first adaptive dynamic correction; K fuzzy is the final fuzzy logic control signal.

[0129] S3: Generate a speed reference control signal based on the proportional coefficient K and the position error e. This signal is used to drive the actuator to control the speed reference value of the system. The speed reference control signal output to the servo system is:

[0130] u=K*e

[0131] Where u is the speed reference control signal; K is the proportional coefficient after the second adaptive dynamic correction; and e is the position error.

[0132] like Figure 5 As shown in the figure, under the same hyperbolic cosine parameters and sinusoidal motion conditions, the left curve shows the error trajectory before fuzzy logic control is introduced, while the right curve shows the error trajectory after fuzzy logic control is introduced. Comparing the system error curves before and after the introduction of fuzzy logic control shows that the system startup error is significantly reduced by 1 mil after the introduction of fuzzy logic control, and the dynamic response performance is significantly improved. Experimental results show that this control strategy effectively improves the system's tracking performance.

[0133] In summary, for the time-varying, strongly coupled, nonlinear servo system, this technical solution uses a hyperbolic cosine function to adjust the proportional coefficient, ensuring both the system's response speed in small error ranges and the required steady-state error once the system is in place. Furthermore, by dynamically optimizing the proportional parameter in conjunction with fuzzy logic control, the system's tracking accuracy and robustness are further enhanced. This solution balances both static and dynamic performance requirements, effectively suppressing overshoot while ensuring response speed and significantly enhancing the system's anti-interference capability.

[0134] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications made to the present invention by those skilled in the art without departing from the spirit of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A control method for a servo system, characterized in that: The small error segment hyperbolic cosine formula is used to perform the first adaptive dynamic correction on the proportional coefficient of the PID controller in the servo system. Based on the proportional coefficient after the first adaptive dynamic correction and the position error of the servo system, the speed given control signal of the servo system is output for real-time control of the servo system.

2. The control method for a servo system according to claim 1, characterized in that: A fuzzy controller is constructed. Before the servo system outputs a speed given control signal, the fuzzy controller performs a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction based on the position error and the position error change rate obtained in real time. Based on the proportional coefficient after the second adaptive dynamic correction and the position error of the servo system, a speed given control signal of the servo system is output for real-time control of the servo system.

3. The control method for a servo system according to claim 1, characterized in that: The small error segment hyperbolic cosine formula is used to perform the first adaptive dynamic correction on the proportional coefficient of the PID controller in the servo system, including the following steps: ① The hyperbolic cosine formula for the small error segment is established as follows: Where K cosh is the proportional coefficient after the first adaptive dynamic correction; a is the factor that determines K cosh The minimum positive integer parameter; b and a together determine K cosh The maximum value is a positive integer parameter; c is a positive integer parameter that determines the rate of change of the hyperbolic cosine curve; e is the position error; and m is a correction coefficient used to adjust the positive integer parameter c. ② According to the requirements of the servo system to be adjusted into position, determine the value range of the proportional coefficient; ③ Adjust parameter b so that the servo system can achieve the fastest response speed when the positioning requirement is met; ④ According to the oscillation of the servo system, adjust the parameter c to improve the stability of the small error segment and output the proportional coefficient of the first adaptive dynamic correction.

4. The control method for a servo system according to claim 2, characterized in that: Before the servo system outputs a speed reference control signal, the steps of performing a second adaptive dynamic correction on the proportional coefficient after the first adaptive dynamic correction based on the position error and the position error change rate obtained in real time include: 1) Obtain position error and position error change rate in real time as input variables of the fuzzy controller; 2) Perform fuzzification processing on each input variable to obtain the fuzzy linguistic variables of each input variable; 3) Construct a fuzzy rule table and use it to infer the fuzzy language variables of the input variables to obtain the fuzzy language of the output variables; 4) Defuzzify the fuzzy language of the output variable to obtain the final fuzzy logic control signal; 5) According to the final fuzzy logic control signal, the proportional coefficient of the second adaptive dynamic correction is output.

5. The control method for a servo system according to claim 4, characterized in that: In step 2), the specific steps of performing fuzzification processing on the input variable to obtain the fuzzy linguistic variable of the input variable include: 2-1) Determine the domain of input variables based on the actual needs of the system; 2-2) Divide the domain of input variables into several fuzzy sets and define corresponding fuzzy linguistic variables; 2-3) Define the membership function of the input variables, and obtain the membership of each input variable on each fuzzy linguistic variable based on the membership function to form the fuzzy linguistic variable of the input variable.

6. The control method for a servo system according to claim 4, characterized in that: In step 4), the fuzzy language of the output variable is defuzzified using the centroid method to obtain the final fuzzy logic control signal. The specific steps include: 4-1) Define the central value of each fuzzy linguistic variable of the output variable; 4-2) Multiply each membership degree of the output variable by the corresponding central value and sum them to obtain a weighted sum; 4-3) Add up all the memberships of the output variables to get the membership sum; 4-4) According to the weighted sum and the membership sum, the final fuzzy logic control signal is obtained.

7. The control method for a servo system according to claim 4, characterized in that: In step 5), the expression for outputting the proportional coefficient of the second adaptive dynamic correction is as follows: K=K cosh +K fuzzy Where, K is the proportional coefficient after the second adaptive dynamic correction; K cosh is the proportional coefficient after the first adaptive dynamic correction; K fuzzy is the final fuzzy logic control signal.

8. The control method for a servo system according to claim 2, characterized in that: The speed setting control signal of the output servo system is: u=K*e Where u is the speed reference control signal; K is the proportional coefficient after the second adaptive dynamic correction; and e is the position error.

Images