AC servo system speed loop parameter self-correction method based on prediction compensation
阅读说明：本技术 一种基于预测补偿的交流伺服系统速度环参数自校正方法 (AC servo system speed loop parameter self-correction method based on prediction compensation ) 是由 卢少武 刘婕 吴波 周凤星 马娅婕 但峰 严保康 胡轶 宁博文 于 2019-10-12 设计创作，主要内容包括：本发明涉及交流伺服控制系统技术领域,提供了一种基于预测补偿的交流伺服系统速度环参数自校正方法,包括S1,对交流伺服系统进行处理,处理包括进行初始化以及进行参数设置；S2,提取处理好后的交流伺服系统的电流与速度,并在线辨识速度环被控模型参数；S3,用未来估计速度与提取的实际速度之间的误差作为补偿,削弱被控模型在线辨识存在的收敛滞后,得到估计速度输出,并以估计速度代替实际速度作为系统反馈；S4,根据估计速度求取速度环最优控制率,并映射二自由度PI控制参数。本发明实现惯量和外部负载扰动大范围变化情况下交流伺服系统速度环控制参数自动校正,工程人员不需要根据手动设定和调节控制器参数,系统自动完成速度环参数自校正。(The invention relates to the technical field of alternating current servo control systems, and provides a predictive compensation-based method for automatically correcting speed loop parameters of an alternating current servo system, which comprises S1, wherein the method comprises the steps of processing the alternating current servo system, and the processing comprises initialization and parameter setting; s2, extracting the current and the speed of the processed alternating current servo system, and identifying the controlled model parameters of the speed loop on line; s3, using the error between the future estimated speed and the extracted actual speed as compensation, weakening the convergence lag existing in the online identification of the controlled model, obtaining the estimated speed output, and using the estimated speed to replace the actual speed as system feedback; and S4, obtaining the optimal control rate of the speed loop according to the estimated speed, and mapping the two-degree-of-freedom PI control parameters. The invention realizes the automatic correction of the speed loop control parameters of the alternating current servo system under the condition of large-range variation of inertia and external load disturbance, and the system automatically finishes the automatic correction of the speed loop parameters without manually setting and adjusting the parameters of the controller by engineering personnel.)
1. A self-correction method for speed loop parameters of an alternating current servo system based on prediction compensation is characterized by comprising the following steps:
s1, processing the alternating current servo system, wherein the processing comprises initialization and parameter setting;
s2, extracting the processed current and speed of the alternating current servo system, and identifying the controlled model parameters of the speed loop on line;
s3, using the error between the future estimated speed and the extracted actual speed as compensation, weakening the convergence lag existing in the online identification of the controlled model, obtaining the estimated speed output, and using the estimated speed to replace the actual speed as system feedback;
and S4, obtaining the optimal control rate of the speed loop according to the estimated speed, and mapping the two-degree-of-freedom PI control parameters to realize the predictive self-correction of the two-degree-of-freedom PI control parameters of the speed loop.
2. The method as claimed in claim 1, wherein in step S2, the parameters of the controlled model of the velocity loop are calculated on line by using recursive least squares algorithm.
3. The method for self-correcting the speed loop parameter of the alternating current servo system based on the predictive compensation as claimed in claim 2, wherein the recursive least square algorithm is as follows:
wherein k isAt the time of sampling,
4. The method for self-correcting the speed loop parameter of the ac servo system based on the predictive compensation as claimed in claim 1, wherein in the step S2, the discrete expression of the speed loop controlled model is:
A(z -1)ω f(k)＝B(z -1)i qr(k-1)
A(z -1)＝1+a 1z -1
B(z -1)＝b 1
wherein, a 1And b 1Is the model parameter, ω, that needs to be identified fTo servo the actual velocity, i qrIs the torque current.
5. The method for self-correcting the velocity loop parameter of an ac servo system based on predictive compensation as claimed in claim 1, wherein in step S3, the compensation is provided by a model error compensator, and the model error compensator employs a PI controller to rapidly attenuate the mismatch error to zero, and the expression can be expressed as:
Δi qm(k)＝k p1[e(k)-e(k-1)]+k i1e(k)
wherein k is p1And k i1Proportional and integral control parameters.
6. The method of claim 1, wherein the method comprises: in step S4, a generalized predictive control is used to find an optimal solution for the speed loop.
7. The method for self-correcting the speed loop parameter of the alternating current servo system based on the predictive compensation as claimed in claim 6, wherein the step of obtaining comprises the following steps:
s40, prediction output: using the Diphantation equation, in combination with the controlled model parameters of the velocity loop, to optimally predict the velocity estimation output at the time k + j, the result can be expressed as,
where Δ is a difference factor, Δ ═ 1-z -1ξ mean zero and variance σ 2White noise of (2);
s41, optimal solution: after obtaining the speed output of the servo system in a certain time period in the future, the control performance of the servo system needs to be evaluated online, and the optimal control quantity of the system is determined according to the evaluation result, which can be expressed as,
where G is the transformation matrix, λ is the control increment weighting coefficient, ω r(k) Is a speed command;
s42, parameter matching: combining the optimal control quantity derived by generalized predictive control and the increment of the two-degree-of-freedom PI controller to obtain the parameter online correction result of the two-degree-of-freedom PI controller, expressing the two-degree-of-freedom PI controller as the following increment mode,
wherein k is p2，k p3Proportional control parameter, k, for a two-degree-of-freedom PI controller i2，k i3Is an integral control parameter.
8. The method for self-correcting the speed loop parameter of the alternating current servo system based on the predictive compensation as claimed in claim 7, wherein in the step S41, the online evaluation is obtained by the following algorithm:
wherein N is 1For minimum prediction length, N is usually taken 1＝1；N 2Is the maximum prediction length; n is a radical of uTo control the time domain length, N is typically used u≤N 2。
9. The method as claimed in claim 7, wherein in step S42, the control parameters of the two-degree-of-freedom PI controller are:
k p2＝-p 1,k i2＝p 1+p 2,k p3＝-f 1,k i3＝f 1+f 2。
Technical Field
The invention relates to the technical field of alternating current servo control systems, in particular to a predictive compensation-based speed loop parameter self-correction method for an alternating current servo system.
Background
The alternating current servo system is an execution unit and a power mechanism of a numerical control machine, is a foundation and a core of the national mechanical manufacturing industry, and is a support of the modern industry in China. With the increase of the demand of the industries of aerospace, automobile, ship, textile and the like on high-grade numerical control machine tools in China, the demand of high-performance alternating current servo systems is larger and larger, but the gap between the control performance and the intelligent level of the current domestic alternating current servo systems and the advanced technology at abroad is still large.
During operation of an ac servo system, the speed command may need to be adjusted, and for such adjustment, the servo drive needs to have good transient response tracking. When the speed instruction is constant, the servo drive needs to have stronger disturbance resistance capability aiming at different operation conditions. Thus, it is difficult for the servo driver to satisfy the requirements of transient response and disturbance resistance simultaneously by using a single-degree-of-freedom PI controller. A large number of researches show that the two-degree-of-freedom PI controller can well solve the problem and can effectively improve the dynamic performance of the alternating current servo system on the basis of not influencing the closed loop stability of the alternating current servo system. However, the two-degree-of-freedom PI controller has more control parameters which need to be adjusted in real time, and in order to meet the development trend of high speed and high precision of an alternating current servo system, an efficient two-degree-of-freedom PI control parameter self-correction strategy needs to be explored.
Generally, control parameter self-calibration methods can be classified into the following two categories: one is a self-calibration method based on rules, such as fuzzy PID, neural network, etc. (X.Duan, H.Deng, H.Li, A failure-based tuning method for fuzzy PID controller [ J ]. IEEE Transactions on Industrial Electronics,2013,60(11): 5177-. Such methods do not rely on accurate mathematical models, but are computationally expensive. The method is suitable for the situation that the controlled object is in impulse response and step response, but is not beneficial to online adjustment with higher real-time requirement. The other type is a self-correcting method based on a model, and an IP control parameter self-correcting method based on generalized Predictive control is proposed in the literature (S.Lu, F.Zhou, Y.Ma, X.Tang, Predictive IP controller for robust position control of linear system [ J ], ISA Transactions,2016,63: 211-. The model-based self-correction method is simple in algorithm and good in stability, but depends on the identification precision of the structure and parameters of the controlled model.
Disclosure of Invention
The invention aims to provide a predictive compensation-based speed loop parameter self-correction method for an alternating current servo system, which can at least solve part of defects in the prior art.
In order to achieve the above purpose, the embodiments of the present invention provide the following technical solutions: a self-correction method for speed loop parameters of an alternating current servo system based on prediction compensation comprises the following steps:
s1, processing the alternating current servo system, wherein the processing comprises initialization and parameter setting;
s2, extracting the processed current and speed of the alternating current servo system, and identifying the controlled model parameters of the speed loop on line;
s3, using the error between the future estimated speed and the extracted actual speed as compensation, weakening the convergence lag existing in the online identification of the controlled model, obtaining the estimated speed output, and using the estimated speed to replace the actual speed as system feedback;
and S4, obtaining the optimal control rate of the speed loop according to the estimated speed, and mapping the two-degree-of-freedom PI control parameters to realize the predictive self-correction of the two-degree-of-freedom PI control parameters of the speed loop.
Further, in the step S2, the velocity-loop controlled model parameters are calculated on line by using a recursive least square algorithm.
Further, the recursive least squares algorithm is as follows:
wherein k is the sampling time,
Further, in the step S2, the discrete expression of the velocity loop controlled model is:
A(z -1)ω f(k)＝B(z -1)i qr(k-1)
A(z -1)＝1+a 1z -1
B(z -1)＝b 1
wherein, a 1And b 1Is the model parameter, ω, that needs to be identified fTo servo the actual velocity, i qrIs the torque current.
Further, in the step S3, the compensation is provided by a model error compensator, which employs a PI controller to rapidly attenuate the mismatch error to zero, and the expression may be expressed as:
Δi qm(k)＝k p1[e(k)-e(k-1)]+k i1e(k)
wherein k is p1And k i1Proportional and integral control parameters.
Further, in the step S4, a generalized predictive control is used to find the optimal solution for the speed loop.
Further, the specific steps of the calculation are as follows:
s40, prediction output: using the Diphantation equation, in combination with the controlled model parameters of the velocity loop, to optimally predict the velocity estimation output at the time k + j, the result can be expressed as,
where Δ is a difference factor, Δ ═ 1-z -1ξ mean zero and variance σ 2White noise of (2);
s41, optimal solution: after obtaining the speed output of the servo system in a certain time period in the future, the control performance of the servo system needs to be evaluated online, and the optimal control quantity of the system is determined according to the evaluation result, which can be expressed as,
where G is the transformation matrix, λ is the control increment weighting coefficient, ω r(k) Is a speed command;
s42, parameter matching: combining the optimal control quantity derived by generalized predictive control and the increment of the two-degree-of-freedom PI controller to obtain the parameter online correction result of the two-degree-of-freedom PI controller, expressing the two-degree-of-freedom PI controller as the following increment mode,
wherein k is p2，k p3Proportional control parameter, k, for a two-degree-of-freedom PI controller i2，k i3Is an integral control parameter.
Further, in the step S41, the online evaluation is obtained by the following algorithm:
wherein N is 1For minimum prediction length, N is usually taken 1＝1；N 2Is the maximum prediction length; n is a radical of uIn order to control the length of the time domain,general formula N u≤N 2。
Further, in the step S42, the control parameters of the two-degree-of-freedom PI controller are:
k p2＝-p 1,k i2＝p 1+p 2,k p3＝-f 1,k i3＝f 1+f 2。
compared with the prior art, the invention has the beneficial effects that:
1. the method can realize the automatic correction of the speed loop control parameters of the alternating current servo system under the condition of large-range variation of inertia and external load disturbance, and the system automatically finishes the automatic correction of the speed loop control parameters without manually setting and adjusting the parameters of the controller by engineering personnel.
2. And a model error compensation loop is established, and the estimated speed output is used as system feedback instead of the actual speed output, so that the dependence degree of the generalized predictive control on the model precision is greatly weakened.
3. Aiming at the unique control structure of the two-degree-of-freedom PI controller, the generalized predictive controller can directly replace the two-degree-of-freedom PI, and meanwhile, the excellent characteristics of the two-degree-of-freedom PI controller are kept.
Drawings
Fig. 1 is a schematic view of a servo system vector control structure of an ac servo system velocity loop parameter self-calibration method based on predictive compensation according to an embodiment of the present invention;
fig. 2 is a structural diagram of a two-degree-of-freedom PI controller of an ac servo system speed loop parameter self-correction method based on predictive compensation according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a control parameter self-correction principle of an ac servo system speed loop parameter self-correction method based on predictive compensation according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating steps of a method for self-correcting a velocity loop parameter of an ac servo system based on predictive compensation according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 4, an embodiment of the present invention provides a method for self-correcting a velocity loop parameter of an ac servo system based on predictive compensation, including the following steps: s1, processing the alternating current servo system, wherein the processing comprises initialization and parameter setting; s2, extracting the processed current and speed of the alternating current servo system, and identifying the controlled model parameters of the speed loop on line; s3, using the error between the future estimated speed and the extracted actual speed as compensation, weakening the convergence lag existing in the online identification of the controlled model, obtaining the estimated speed output, and using the estimated speed to replace the actual speed as system feedback; and S4, obtaining the optimal control rate of the speed loop according to the estimated speed, and mapping the two-degree-of-freedom PI control parameters to realize the predictive self-correction of the two-degree-of-freedom PI control parameters of the speed loop. According to the embodiment, the automatic correction of the speed loop control parameters of the alternating current servo system under the condition of large-range variation of inertia and external load disturbance can be realized, and the system automatically finishes the automatic correction of the speed loop control parameters without manually setting and adjusting the parameters of the controller by an engineer.
Referring to FIG. 1, in practical engineering applications, i is usually adopted dApproximate decoupling of the currents is achieved as 0. Under the vector control structure, the speed loop controlled object model of the servo system can be expressed by the following formula:
wherein, ω is fAs the actual speed, i qrIs a moment current, k fFor the moment coefficient, T, associated with the flux linkage of the machine pIs the inertia constant, J is the system moment of inertia, f distFor external loads of the servo system, τ dFor serving systemsThe dead time of (d). Discretizing the equation (1) can obtain a first-order discrete model of the controlled object of the speed ring:
A(z -1)ω f(k)＝B(z -1)i qr(k-1) (2)
A(z -1)＝1+a 1z -1(3)
B(z -1)＝b 1(4)
wherein, a 1And b 1Are the model parameters that need to be identified.
The model error compensation loop employs a PI controller whose incremental discrete mode can be expressed as:
Δi qm(k)＝k p1[e(k)-e(k-1)]+k i1e(k) (6)
wherein k is p1And k i1For proportional and integral control parameters, Δ is a difference factor, Δ ═ 1-z -1。
The speed control loop adopts a two-degree-of-freedom PI controller, and the increment discrete mode of the speed control loop can be expressed as follows:
wherein k is p2，k p3Proportional control parameter, k, for a two-degree-of-freedom PI controller i2，k i3For integrating the control parameter, ω r(k) In order to be a speed command,
Referring to fig. 2 and 3, after obtaining the controlled model parameters of the speed loop, a model error compensation loop is established to reduce the model mismatching error, calculate the estimated speed output, and use the estimated speed to replace the actual speed as the system feedback; meanwhile, in a speed control loop, the estimated speed output at the future moment is optimally predicted, a Diphanine equation is simplified, the optimal control rate of a servo system is obtained according to the evaluation result, and the self-correction of the control parameters of the speed loop two-degree-of-freedom PI controller is realized. Mainly comprises the following steps:
first, the real speed ω in the servo system speed loop needs to be extracted in real time fAnd current i qr. As input and output data of the recursive least squares algorithm. Obtaining the required controlled model parameter a through real-time online identification 1And b 1The recursive least squares algorithm is as follows:
wherein k is the sampling time,
And secondly, after the model parameters are obtained, the estimated speed output can be calculated, and a model error compensation loop is established. The model error compensation method adopts the error between the actual output and the estimated output as the compensation quantity, and weakens the degree of dependence of the generalized predictive control on the model precision. Thus, the dynamic equation for the model error compensation is:
i qe(k)＝i qr(k)+i qm(k) (10)
wherein i qm(k) Is a model error compensation quantity.
And thirdly, after obtaining the model parameters and the speed estimation output, obtaining the prediction output of the estimated speed at the k + j moment according to a simplified Diophantine equation. The estimated speed prediction output result can be expressed as:
wherein ξ represents a mean value of zero and a variance of σ 2White noise of (2).
In a speed control loop, the speed estimation output is used as system feedback, a speed command is a step signal, and the performance indexes of a speed loop of an alternating current servo system are established as follows:
wherein, ω is rFor speed command, N 1For minimum prediction length, N is usually taken 1＝1；N 2Is the maximum prediction length; n is a radical of uTo control the time domain length, N is typically used u≤N 2(ii) a λ is the control increment weighting coefficient.
By substituting the estimated velocity prediction output in equation (12) into equation (13), in order to minimize the control performance index represented by equation (6), the optimum control amount of the servo system position loop can be obtained by using the principles of roll optimization and feedback correction, which can be represented as:
get N 2Substitution of formula (14) may yield 2:
comparing the formula (7) with the formula (15), the control parameters of the two-degree-of-freedom PI controller can be obtained:
k p2＝-p 1,k i2＝p 1+p 2,k p3＝-f 1,k i3＝f 1+f 2(16)
although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.