CN113467236B - A method of time delay compensation for error signal - Google Patents

A method of time delay compensation for error signal Download PDF

Info

Publication number
CN113467236B
CN113467236B CN202110671903.2A CN202110671903A CN113467236B CN 113467236 B CN113467236 B CN 113467236B CN 202110671903 A CN202110671903 A CN 202110671903A CN 113467236 B CN113467236 B CN 113467236B
Authority
CN
China
Prior art keywords
error signal
time
delay
channel
estimation model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN202110671903.2A
Other languages
Chinese (zh)
Other versions
CN113467236A (en
Inventor
马召召
高伟鹏
杨庆超
柴凯
刘树勇
楼京俊
周瑞平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan University of Technology WUT
Naval University of Engineering PLA
Original Assignee
Wuhan University of Technology WUT
Naval University of Engineering PLA
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan University of Technology WUT, Naval University of Engineering PLA filed Critical Wuhan University of Technology WUT
Priority to CN202110671903.2A priority Critical patent/CN113467236B/en
Publication of CN113467236A publication Critical patent/CN113467236A/en
Application granted granted Critical
Publication of CN113467236B publication Critical patent/CN113467236B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/12—Timing analysis or timing optimisation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a method for time lag compensation of error signals, which comprises the following steps: 1) Identifying a time-lag channel in a control system; identifying a primary channel and a secondary channel in a control system to obtain a primary channel estimation model P (z) and a secondary channel estimation model
Figure DDA0003119689720000011
2) Collecting controller output signal y (n) and error signal e in control process t (n) combining the channel estimation model to identify the time-lag channel filter on line to obtain a time-lag transfer function estimation model
Figure DDA0003119689720000012
3) According to
Figure DDA0003119689720000013
Calculating an inverse system B (z) and using it on the acquired error signal e t (n) inverse filtering to obtain the estimated value of the real error signal
Figure DDA0003119689720000014
4) According to the collected error signal e t (n) and true error signal estimate
Figure DDA0003119689720000015
Establishing a moving average model, and correcting the estimation error signal to obtain
Figure DDA0003119689720000016
And the error signal is used as a real error signal after time lag compensation and is put into a control system to control the iterative calculation of filter parameters. The method can improve the convergence speed and the control precision of the main control link.

Description

一种对误差信号进行时滞补偿的方法A method of time delay compensation for error signal

技术领域technical field

本发明属于数字信号处理技术领域，具体涉及一种对误差信号进行时滞补偿的方法。The invention belongs to the technical field of digital signal processing, and in particular relates to a method for performing time delay compensation on an error signal.

背景技术Background technique

变换域块最小均方算法实时性较差，时滞的存在会影响算法的收敛性和稳定性，即使自适应算法对时滞有一定的适应能力，仍需对系统时滞进行补偿。时滞的来源主要有三个方面：一、信号在硬件之间的采集、传输过程；二、对采集信号进行预处理、参数迭代更新的运算过程；三、输出信号至误差信号传感器采集信号的过程。第一部分由系统硬件性能决定；第二部分由控制处理器性能和算法具体实现运算量和过程相关；第三部分引起的时滞包含在次级通道的相位中，估计模型

Figure BDA0003119689700000011
对参考信号的滤波可视为对此时滞的补偿。The transform domain block least mean square algorithm has poor real-time performance, and the existence of time delay will affect the convergence and stability of the algorithm. Even if the adaptive algorithm has a certain ability to adapt to the time delay, it still needs to compensate for the system time delay. There are three main sources of time delay: 1. The process of signal acquisition and transmission between hardware; 2. The operation process of preprocessing the acquired signal and iteratively updating parameters; 3. The process of outputting the signal to the error signal sensor to collect the signal . The first part is determined by the system hardware performance; the second part is related to the performance of the control processor and the specific implementation of the algorithm and the calculation amount and process; the third part of the time delay caused by
Figure BDA0003119689700000011
Filtering of the reference signal can be viewed as compensation for this skew.

在较为复杂的控制算法中，参考信号到输出信号之间经过复杂的滤波运算，造成的时滞对控制效果的影响较大，甚至出现控制发散的情况，对时滞进行在线辨识和补偿是非常有必要的。以往研究中，补偿时滞对控制系统的影响大体分为两种：1)将随机时滞转化为固定时延，设计控制器；2)对随机时延建模分析，在算法进行补偿。前者虽结构简单，但将随机时滞转化为固定时滞时，存在精度不高的问题，本申请中提出逆建模控制算法，通过时间序列预估补偿后的误差信号，用于整个控制系统输出力的运算。In a more complex control algorithm, the time delay caused by the complex filtering operation between the reference signal and the output signal has a great influence on the control effect, and even the control divergence occurs. It is very important to identify and compensate the time delay online. Necessary. In previous studies, the effects of compensation delay on the control system can be roughly divided into two types: 1) Convert random delay into fixed delay and design the controller; 2) Model and analyze the random delay and compensate in the algorithm. Although the former has a simple structure, it has the problem of low precision when converting random time delays into fixed time delays. In this application, an inverse modeling control algorithm is proposed to estimate the compensated error signal through time series, which is used for the entire control system. Calculation of output force.

发明内容SUMMARY OF THE INVENTION

本发明要解决的技术问题在于针对现有技术存在的不足，提供一种对误差进行时滞补偿的方法，可显著提高收敛速度和控制精度。The technical problem to be solved by the present invention is to provide a method for time delay compensation for errors, which can significantly improve the convergence speed and control accuracy, aiming at the shortcomings of the prior art.

本发明解决其技术问题所采用的技术方案是：The technical scheme adopted by the present invention to solve its technical problems is:

一种对误差信号进行时滞补偿的方法，包括如下步骤：A method for performing time delay compensation on an error signal, comprising the following steps:

1)控制系统中时滞通道的辨识；对控制系统中初级通道和次级通道进行辨识，得初级通道估计模型P(z)和次级通道估计模型

Figure BDA0003119689700000012
1) Identification of the delay channel in the control system; identify the primary channel and the secondary channel in the control system to obtain the primary channel estimation model P(z) and the secondary channel estimation model
Figure BDA0003119689700000012

2)采集控制过程中的控制器输出信号y(n)和误差信号et(n)，结合通道估计模型在线辨识时滞通道滤波器，得到时滞传递函数估计模型

Figure BDA0003119689700000021
2) Collect the controller output signal y(n) and the error signal e t (n) in the control process, identify the time-delay channel filter online with the channel estimation model, and obtain the time-delay transfer function estimation model
Figure BDA0003119689700000021

3)根据

Figure BDA0003119689700000022
计算逆系统B(z)，并用之对采集的误差信号et(n)进行逆滤波，得到真实误差信号的估计值
Figure BDA0003119689700000023
3) According to
Figure BDA0003119689700000022
Calculate the inverse system B(z), and use it to inverse filter the collected error signal e t (n) to obtain the estimated value of the real error signal
Figure BDA0003119689700000023

4)实际控制系统当中，时滞直观体现在时间序列上，为进一步提高误差信号估计精度，根据采集的误差信号et(n)和真实误差信号估计值

Figure BDA0003119689700000024
建立滑动平均(MA)模型，基于时间序列模型对估计误差信号进行修正，得到
Figure BDA0003119689700000025
将其作为时滞补偿后的真实的误差信号，放入控制系统进行控制滤波器参数迭代计算。4) In the actual control system, the time delay is directly reflected in the time series. In order to further improve the estimation accuracy of the error signal, according to the collected error signal e t (n) and the estimated value of the real error signal
Figure BDA0003119689700000024
A moving average (MA) model is established, and the estimated error signal is corrected based on the time series model to obtain
Figure BDA0003119689700000025
Take it as the real error signal after time delay compensation, and put it into the control system for iterative calculation of control filter parameters.

按上述方案，所述初级通道估计模型P(z)和次级通道估计模型

Figure BDA0003119689700000026
通过辨识算法得出。According to the above scheme, the primary channel estimation model P(z) and the secondary channel estimation model
Figure BDA0003119689700000026
obtained through the identification algorithm.

按上述方案，所述步骤2)

Figure BDA0003119689700000027
的更新公式为According to the above scheme, the step 2)
Figure BDA0003119689700000027
The update formula for is

Figure BDA0003119689700000028
Figure BDA0003119689700000028

其中，d(n)为时滞通道辨识环节的输入信号，ht(n)为时滞辨识滤波器的误差信号，μ为迭代步长。Among them, d(n) is the input signal of the time-delay channel identification link, h t (n) is the error signal of the time-delay identification filter, and μ is the iterative step size.

按上述方案，所述步骤4)中还包括对真实误差信号的估计值

Figure BDA0003119689700000029
修正的步骤，修正方法如下：According to the above scheme, the step 4) also includes the estimated value of the real error signal
Figure BDA0003119689700000029
The correction steps are as follows:

依据滑动平均模型和最小二乘递推算法得到加权系数，预测出n时刻的真实误差信号

Figure BDA00031196897000000210
将其用作对
Figure BDA00031196897000000211
的修正：According to the moving average model and the least square recursive algorithm, the weighting coefficient is obtained, and the real error signal at time n is predicted.
Figure BDA00031196897000000210
use it as a pair
Figure BDA00031196897000000211
Correction:

根据采集的误差信号et(n)和真实误差信号的估计值

Figure BDA00031196897000000212
建立滑动平均模型，数学表达式为According to the estimated value of the collected error signal e t (n) and the real error signal
Figure BDA00031196897000000212
A moving average model is established, and the mathematical expression is

Figure BDA00031196897000000213
Figure BDA00031196897000000213

式中，ai、ci为MA系数，M、N为MA模型阶，

Figure BDA00031196897000000214
为修正后真实的误差信号；where a i and c i are MA coefficients, M and N are MA model orders,
Figure BDA00031196897000000214
is the real error signal after correction;

将其表示为最小二乘形式Express it in least squares form

Figure BDA0003119689700000031
Figure BDA0003119689700000031

其中in

Figure BDA0003119689700000032
Figure BDA0003119689700000032

θ＝[a1,a2,...aM,c0,c1,...,cN]T θ=[a 1 ,a 2 ,...a M ,c 0 ,c 1 ,...,c N ] T

使用带遗忘因子的递推最小二乘法进行参数辨识得到θ，以步骤3)得到的估计误差

Figure BDA0003119689700000033
为目标函数，使得
Figure BDA0003119689700000034
逼近
Figure BDA0003119689700000035
并对其进行修正，递推算法为Use the recursive least squares method with forgetting factor for parameter identification to obtain θ, and use the estimated error obtained in step 3).
Figure BDA0003119689700000033
is the objective function, so that
Figure BDA0003119689700000034
Approach
Figure BDA0003119689700000035
and modify it, the recursive algorithm is

Figure BDA0003119689700000036
Figure BDA0003119689700000036

其中，δ为遗忘因子。where δ is the forgetting factor.

本发明产生的有益效果是：本发明方法能高精度逼近真实的时滞通道；时滞预测补偿可真实的还原出误差信号，进一步提高了主控制环节的收敛速度和控制精度。The beneficial effects of the invention are: the method of the invention can approach the real time delay channel with high precision; the time delay prediction compensation can truly restore the error signal, which further improves the convergence speed and control precision of the main control link.

附图说明Description of drawings

图1为本发明实施例的考虑时滞的控制系统的结构图；1 is a structural diagram of a control system considering time delay according to an embodiment of the present invention;

图2为本发明实施例的时滞在线估计方法示意图；FIG. 2 is a schematic diagram of a time delay online estimation method according to an embodiment of the present invention;

图3为本发明实施例的误差信号时滞补偿过程示意图。FIG. 3 is a schematic diagram of an error signal skew compensation process according to an embodiment of the present invention.

具体实施方式Detailed ways

下面申请人将结合具体的实施例对本发明作进一步的详细说明，以使本领域的技术人员更加清楚地理解本发明。但以下内容不应理解为是对本发明的权利要求书请求保护范围的限制。The applicant will further describe the present invention in detail with reference to specific embodiments, so that those skilled in the art can understand the present invention more clearly. However, the following contents should not be construed as limiting the scope of protection of the claims of the present invention.

如图1所示，时滞的存在对控制算法会造成一定的影响，将时滞对控制系统的影响看作是一段传递路径，传递函数为T(z)，幅值基本不变，相位随频率变化。激励频率波动时，时滞为一变化参数，无法直接测量得出，需对T(z)进行辨识。在较为复杂的控制算法中，参考信号到输出信号之间经过复杂的滤波运算，造成的时滞对控制效果的影响较大，甚至出现控制发散的情况，对时滞进行在线辨识和补偿是非常有必要的。As shown in Figure 1, the existence of time delay will have a certain impact on the control algorithm. The impact of time delay on the control system is regarded as a transfer path. The transfer function is T(z), the amplitude is basically unchanged, and the phase varies with frequency changes. When the excitation frequency fluctuates, the time lag is a variable parameter, which cannot be directly measured, and T(z) needs to be identified. In a more complex control algorithm, the time delay caused by the complex filtering operation between the reference signal and the output signal has a great influence on the control effect, and even the control divergence occurs. It is very important to identify and compensate the time delay online. Necessary.

一种对误差信号进行时滞补偿的方法，包括如下步骤：A method for performing time delay compensation on an error signal, comprising the following steps:

1)控制系统中时滞通道的辨识；对控制系统中初级通道和次级通道进行辨识，得初级通道估计模型P(z)和次级通道估计模型

Figure BDA0003119689700000041
1) Identification of the delay channel in the control system; identify the primary channel and the secondary channel in the control system to obtain the primary channel estimation model P(z) and the secondary channel estimation model
Figure BDA0003119689700000041

如图2，根据主动控制系统内的时滞产生特点可知，进入控制器的误差信号实际上是经过了时滞通道的误差信号et(n)，其表达式见式(1)：As shown in Figure 2, according to the time-delay generation characteristics in the active control system, the error signal entering the controller is actually the error signal e t (n) that has passed through the time-delay channel, and its expression is shown in formula (1):

et(n)＝e(n)T(z)＝d(n)T(z)+y(n)S(z)T(z) (1)e t (n)=e(n)T(z)=d(n)T(z)+y(n)S(z)T(z) (1)

经过时滞通道的误差信号由两部分组成，分别为经过初级时滞通道的期望信号d(n)和经过次级通道滤波的输出信号y(n)S(z)，其中d(n)为参考信号x(n)经过初级通道P(z)得到的，y(n)S(z)为输出信号y(n)经过次级通道S(z)得到的；可将其中任一部分作为通过时滞通道后的信号，只需找到其对应的未经过时滞通道的信号分量，通过自适应滤波器即可得到时滞环节的滤波器权系数。The error signal through the time-delay channel consists of two parts, which are the expected signal d(n) through the primary time-delay channel and the output signal y(n)S(z) filtered by the secondary channel, where d(n) is The reference signal x(n) is obtained by passing through the primary channel P(z), and y(n)S(z) is obtained by the output signal y(n) passing through the secondary channel S(z); The signal after the delay channel only needs to find the corresponding signal component of the channel without delay, and then the filter weight coefficient of the delay link can be obtained through the adaptive filter.

在实际控制系统中，初级通道只与双层隔振平台的系统特性相关，且一般不会发生变化；次级通道与作动器输出特性和系统特性相关，非线性影响因素较强，即便通过次级通道辨识算法进行辨识仍存在一定的辨识误差，且随着工况的变化，次级通道特性或会发生变化。从计算复杂度和工程实际应用来看，确定d(n)作为时滞辨识环节的参考输入信号，d(n)T(z)作为时滞辨识环节的目标信号，见式(2)，通过自适应滤波器辨识迭代得到时滞通道的滤波器权系数。In the actual control system, the primary channel is only related to the system characteristics of the double-layer vibration isolation platform, and generally does not change; the secondary channel is related to the output characteristics of the actuator and the system characteristics, and the nonlinear influencing factors are strong. There is still a certain identification error in the identification algorithm of the secondary channel, and with the change of the working conditions, the characteristics of the secondary channel may change. From the perspective of computational complexity and practical engineering application, d(n) is determined as the reference input signal of the time-delay identification process, and d(n)T(z) is used as the target signal of the time-delay identification process. See equation (2). The adaptive filter identification iteratively obtains the filter weight coefficients of the delay channel.

d(n)T(z)＝et(n)-y(n)S(z)T(z) (2)d(n)T(z)=e t (n)-y(n)S(z)T(z) (2)

为得到d(n)T(z)，首先应构造出经过次级通道滤波的输出信号y(n)S(z)，在实际的控制系统中，次级通道S(z)并不是严格可知的，只能用辨识模型

Figure BDA0003119689700000042
来代替。In order to obtain d(n)T(z), the output signal y(n)S(z) filtered by the secondary channel should be constructed first. In the actual control system, the secondary channel S(z) is not strictly known. , only the identification model can be used
Figure BDA0003119689700000042
instead.

2)采集控制过程中的控制器输出信号y(n)和误差信号ed(n)，结合通道估计模型在线辨识时滞通道滤波器，得到时滞传递函数估计模型

Figure BDA0003119689700000051
2) Collect the controller output signal y (n) and the error signal ed (n) in the control process, identify the time-delay channel filter online with the channel estimation model, and obtain the time-delay transfer function estimation model
Figure BDA0003119689700000051

将时滞通道T(z)用时滞环节辨识滤波器权系数

Figure BDA0003119689700000052
表示，
Figure BDA0003119689700000053
看作T(z)的估计模型，以d(n)作为时滞通道辨识环节的输入信号，
Figure BDA0003119689700000054
作为目标信号，以ht(n)作为时滞辨识滤波器的误差信号，则Use the time-delay channel T(z) to identify the filter weight coefficients
Figure BDA0003119689700000052
express,
Figure BDA0003119689700000053
As the estimation model of T(z), d(n) is used as the input signal of the time-delay channel identification link,
Figure BDA0003119689700000054
As the target signal, with h t (n) as the error signal of the time delay identification filter, then

Figure BDA0003119689700000055
Figure BDA0003119689700000055

在不考虑次级通道辨识误差的情况下，随着时滞滤波器权系数的迭代，

Figure BDA0003119689700000056
不断逼近实际的时滞通道T(z)，时滞辨识误差信号ht(n)趋于零，时滞辨识环节完成。仍使用最小均方算法的收敛准则，进行时滞滤波器权系数的迭代：Without considering the secondary channel identification error, with the iteration of the delay filter weight coefficients,
Figure BDA0003119689700000056
Continuously approaching the actual time delay channel T(z), the time delay identification error signal h t (n) tends to zero, and the time delay identification link is completed. Still using the convergence criterion of the least mean squares algorithm, iterate the weight coefficients of the delay filter:

Figure BDA0003119689700000057
Figure BDA0003119689700000057

式中，d(n)＝x(n)P(z)。因此，在进行时滞通道在线辨识前，需对初级通道P(z)进行辨识，并将辨识结果放入时滞辨识中即可。In the formula, d(n)=x(n)P(z). Therefore, before the online identification of the time-delay channel, it is necessary to identify the primary channel P(z), and put the identification result into the time-delay identification.

3)根据

Figure BDA0003119689700000058
计算逆系统B(z)，并用之对采集的误差信号ed(n)进行逆滤波，得到真实误差信号的估计值
Figure BDA0003119689700000059
3) According to
Figure BDA0003119689700000058
Calculate the inverse system B( z ), and use it to inverse filter the collected error signal ed (n) to obtain the estimated value of the real error signal
Figure BDA0003119689700000059

时滞补偿算法如图3，采集得到的误差信号可看作真实的误差信号经过了时滞通道滤波，为还原真实的误差信号，在其进入控制算法前进行逆滤波，所使用的滤波器B(z)应满足：The time-delay compensation algorithm is shown in Figure 3. The collected error signal can be regarded as the real error signal that has been filtered by the time-delay channel. In order to restore the real error signal, inverse filtering is performed before it enters the control algorithm. The filter B used (z) shall satisfy:

Figure BDA00031196897000000510
Figure BDA00031196897000000510

式中，用时滞估计滤波器

Figure BDA00031196897000000511
代替真实的时滞通道。即where the delay estimation filter is used
Figure BDA00031196897000000511
Instead of a real time-delayed channel. which is

Figure BDA00031196897000000512
Figure BDA00031196897000000512

对et(n)进行逆滤波，得到实际误差信号e(n)的估计值Perform inverse filtering on e t (n) to obtain an estimate of the actual error signal e(n)

Figure BDA0003119689700000061
Figure BDA0003119689700000061

由式(4.2.11)即可得到真实误差信号的估计值。在实际的控制系统当中，时滞可直观体现在时间序列上，n时刻的误差信号e(n)在n+τ时刻被采集得到。The estimated value of the real error signal can be obtained by formula (4.2.11). In the actual control system, the time delay can be directly reflected in the time series, and the error signal e(n) at time n is collected at time n+τ.

4)依据MA模型和最小二乘递推算法得到加权系数，预测出n时刻的真实误差信号

Figure BDA0003119689700000062
将其用作对
Figure BDA0003119689700000063
的修正，将得到的
Figure BDA0003119689700000064
作为真实的误差信号，放入控制系统进行控制滤波器参数迭代计算。4) According to the MA model and the least square recursive algorithm, the weighting coefficient is obtained, and the real error signal at time n is predicted.
Figure BDA0003119689700000062
use it as a pair
Figure BDA0003119689700000063
correction, will get the
Figure BDA0003119689700000064
As the real error signal, it is put into the control system for iterative calculation of control filter parameters.

为进一步提高误差信号的估计精度，基于时间序列模型对估计误差信号进行修正。In order to further improve the estimation accuracy of the error signal, the estimated error signal is corrected based on the time series model.

根据采集的误差信号et(n)和真实误差信号的估计值

Figure BDA0003119689700000065
建立滑动平均(MA)模型，数学表达式可写为According to the estimated value of the collected error signal e t (n) and the real error signal
Figure BDA0003119689700000065
To establish a moving average (MA) model, the mathematical expression can be written as

Figure BDA0003119689700000066
Figure BDA0003119689700000066

式中，ai、ci为MA系数，M、N为模型阶数(取正整数)，

Figure BDA0003119689700000067
为修正后真实的误差信号。In the formula, a i and c i are MA coefficients, M and N are model orders (take positive integers),
Figure BDA0003119689700000067
is the corrected real error signal.

将其表示为最小二乘形式Express it in least squares form

Figure BDA0003119689700000068
Figure BDA0003119689700000068

其中in

Figure BDA0003119689700000069
Figure BDA0003119689700000069

θ＝[a1,a2,...aM,c0,c1,...,cN]T θ=[a 1 ,a 2 ,...a M ,c 0 ,c 1 ,...,c N ] T

使用带遗忘因子的递推最小二乘法进行参数辨识得到θ，以式(7)得到的估计误差为目标函数，使得

Figure BDA00031196897000000610
逼近
Figure BDA00031196897000000611
并对其进行修正。递推算法为Use the recursive least squares method with forgetting factor to perform parameter identification to obtain θ, and take the estimated error obtained by equation (7) as the objective function, so that
Figure BDA00031196897000000610
Approach
Figure BDA00031196897000000611
and amend it. The recursive algorithm is

Figure BDA0003119689700000071
Figure BDA0003119689700000071

式中，δ为遗忘因子，一般取值范围为0.96＜δ＜0.99。In the formula, δ is the forgetting factor, and the general value range is 0.96<δ<0.99.

上述实施例只为说明本发明的技术构思及特点，其目的在于让熟悉此项技术的人士能够了解本发明的内容并据以实施，并不能以此限制本发明的保护范围。凡根据本发明精神实质所作的等效变化或修饰，都应涵盖在本发明的保护范围之内。The above-mentioned embodiments are only intended to illustrate the technical concept and characteristics of the present invention, and the purpose thereof is to enable those who are familiar with the art to understand the content of the present invention and implement them accordingly, and cannot limit the protection scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention should be included within the protection scope of the present invention.

Claims (3)

1.一种对误差信号进行时滞补偿的方法，其特征在于，包括如下步骤：1. a method for carrying out time delay compensation to an error signal, is characterized in that, comprises the steps: 1)控制系统中时滞通道的辨识；对控制系统中初级通道和次级通道进行辨识，得初级通道估计模型P(z)和次级通道估计模型
Figure FDA0003719353840000011
1) Identification of the delay channel in the control system; identify the primary channel and the secondary channel in the control system to obtain the primary channel estimation model P(z) and the secondary channel estimation model
Figure FDA0003719353840000011
 2)采集控制过程中的控制器输出信号y(n)和误差信号et(n)，结合通道估计模型在线辨识时滞通道滤波器，得到时滞传递函数估计模型
Figure FDA0003719353840000012
2) Collect the controller output signal y(n) and the error signal e t (n) in the control process, identify the time-delay channel filter online with the channel estimation model, and obtain the time-delay transfer function estimation model
Figure FDA0003719353840000012
Figure FDA0003719353840000013
的更新公式为：
Figure FDA0003719353840000013
The update formula is:
Figure FDA0003719353840000014
Figure FDA0003719353840000014
 其中，d(n)为时滞通道辨识环节的输入信号，ht(n)为时滞辨识滤波器的误差信号，μ为迭代步长；Among them, d(n) is the input signal of the time-delay channel identification link, h t (n) is the error signal of the time-delay identification filter, and μ is the iterative step size; 3)根据
Figure FDA0003719353840000015
计算逆系统B(z)，并用之对采集的误差信号et(n)进行逆滤波，得到真实误差信号的估计值
Figure FDA0003719353840000016
3) According to
Figure FDA0003719353840000015
Calculate the inverse system B(z), and use it to inverse filter the collected error signal e t (n) to obtain the estimated value of the real error signal
Figure FDA0003719353840000016
 4)根据采集的误差信号et(n)和真实误差信号估计值
Figure FDA0003719353840000017
建立滑动平均模型，对估计误差信号进行修正，得到
Figure FDA0003719353840000018
将其作为时滞补偿后的真实的误差信号，放入控制系统进行控制滤波器参数迭代计算。4) According to the collected error signal e t (n) and the estimated value of the real error signal
Figure FDA0003719353840000017
A moving average model is established, and the estimated error signal is corrected to obtain
Figure FDA0003719353840000018
Take it as the real error signal after time delay compensation, and put it into the control system for iterative calculation of control filter parameters. 2.根据权利要求1所述的对误差信号进行时滞补偿的方法，其特征在于，所述初级通道估计模型P(z)和次级通道估计模型
Figure FDA0003719353840000019
通过辨识算法得出。2. The method for performing time delay compensation on an error signal according to claim 1, wherein the primary channel estimation model P(z) and the secondary channel estimation model are
Figure FDA0003719353840000019
obtained through the identification algorithm. 3.根据权利要求1所述的对误差信号进行时滞补偿的方法，其特征在于，所述步骤4)中对估计误差信号进行修正，修正方法如下：3. the method for carrying out time delay compensation to the error signal according to claim 1, is characterized in that, in described step 4), the estimated error signal is corrected, and the correction method is as follows: 依据滑动平均模型和最小二乘递推算法得到加权系数，预测出n时刻的真实误差信号
Figure FDA0003719353840000021
将其用作对
Figure FDA0003719353840000022
的修正：According to the moving average model and the least square recursive algorithm, the weighting coefficient is obtained, and the real error signal at time n is predicted.
Figure FDA0003719353840000021
use it as a pair
Figure FDA0003719353840000022
Correction: 根据采集的误差信号et(n)和真实误差信号的估计值
Figure FDA0003719353840000023
建立滑动平均模型，数学表达式为According to the estimated value of the collected error signal e t (n) and the real error signal
Figure FDA0003719353840000023
A moving average model is established, and the mathematical expression is
Figure FDA0003719353840000024
Figure FDA0003719353840000024
 式中，ai、ci为MA系数，M、N为MA模型阶，
Figure FDA0003719353840000025
为修正后真实的误差信号；where a i and c i are MA coefficients, M and N are MA model orders,
Figure FDA0003719353840000025
is the real error signal after correction; 将其表示为最小二乘形式Express it in least squares form
Figure FDA0003719353840000026
Figure FDA0003719353840000026
 其中in
Figure FDA0003719353840000027
Figure FDA0003719353840000027
 θ＝[a1,a2,...aM,c0,c1,...,cN]T θ=[a 1 ,a 2 ,...a M ,c 0 ,c 1 ,...,c N ] T 使用带遗忘因子的递推最小二乘法进行参数辨识得到θ，以步骤3)得到的估计误差
Figure FDA0003719353840000028
为目标函数，使得
Figure FDA0003719353840000029
逼近
Figure FDA00037193538400000210
并对其进行修正，递推算法为Use the recursive least squares method with forgetting factor for parameter identification to obtain θ, and use the estimated error obtained in step 3).
Figure FDA0003719353840000028
is the objective function, so that
Figure FDA0003719353840000029
Approach
Figure FDA00037193538400000210
and modify it, the recursive algorithm is
Figure FDA00037193538400000211
Figure FDA00037193538400000211
 其中，δ为遗忘因子。where δ is the forgetting factor.
CN202110671903.2A 2021-06-17 2021-06-17 A method of time delay compensation for error signal Expired - Fee Related CN113467236B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110671903.2A CN113467236B (en) 2021-06-17 2021-06-17 A method of time delay compensation for error signal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110671903.2A CN113467236B (en) 2021-06-17 2021-06-17 A method of time delay compensation for error signal

Publications (2)

Publication Number Publication Date
CN113467236A CN113467236A (en) 2021-10-01
CN113467236B true CN113467236B (en) 2022-10-21

Family

ID=77870380

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110671903.2A Expired - Fee Related CN113467236B (en) 2021-06-17 2021-06-17 A method of time delay compensation for error signal

Country Status (1)

Country Link
CN (1) CN113467236B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114459712B (en) * 2022-01-11 2023-12-29 东南大学 Earthquake simulation vibrating table experimental compensation method based on autoregressive model
CN114900401B (en) * 2022-03-24 2024-09-20 重庆邮电大学 DFMA-PONs-oriented channel interference elimination method and device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101924533A (en) * 2010-07-19 2010-12-22 浙江工业大学 Multivariate time-delay parameter estimation method based on FIR model identification
KR20110062291A (en) * 2009-12-03 2011-06-10 한국과학기술원 Skew Estimation Time Delay Control Method of Robot Manipulator and Robot Manipulator Controller
JP2015197941A (en) * 2014-04-03 2015-11-09 ヤマハ株式会社 Sampling frequency estimation device

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1945470A (en) * 2006-11-02 2007-04-11 上海交通大学 Two freedom decoupling smith pre-evaluating control system of industrial multiple variable time lag process
CN101286044B (en) * 2008-05-12 2010-06-16 杭州电子科技大学 A steam temperature mixing modeling method for coal-fired boiler system
CN105467844A (en) * 2016-01-22 2016-04-06 陈昊哲 Boiler overheating steam temperature control method based on Neuron identification

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20110062291A (en) * 2009-12-03 2011-06-10 한국과학기술원 Skew Estimation Time Delay Control Method of Robot Manipulator and Robot Manipulator Controller
CN101924533A (en) * 2010-07-19 2010-12-22 浙江工业大学 Multivariate time-delay parameter estimation method based on FIR model identification
JP2015197941A (en) * 2014-04-03 2015-11-09 ヤマハ株式会社 Sampling frequency estimation device

Also Published As

Publication number Publication date
CN113467236A (en) 2021-10-01

Similar Documents

Publication Publication Date Title
CN113467236B (en) A method of time delay compensation for error signal
CN104009734B (en) Gradient variable step-size LMS adaptive filter method
CN103383261B (en) A kind of modified can&#39;t harm the indoor moving targets location method of Kalman filtering
CN104796111B (en) It is a kind of to be used for Dynamic Hysteresis system modelling and the nonlinear adaptable filter of compensation
CN104238366B (en) The forecast Control Algorithm and device of piezoelectric ceramic actuator based on neuroid
CN111859795B (en) Polynomial aided neural network behavior modeling system and method for power amplifier
CN104270055A (en) Parameter self-correcting method of position loop IP controller of alternating current servo system
CN107272421A (en) A kind of Active Disturbance Rejection Control system and method
CN106411290A (en) Variable step size VLMP filtering algorithm based on sigmoid function and application thereof
CN105652662A (en) Active control method for vibration of piezoelectric structure for adaptive narrow-band filtering
CN108828950A (en) A kind of adaptive Auto-disturbance-rejection Control, device and equipment
CN103941589A (en) Non-linear model predictive control method of piezoelectric actuator
Liu et al. Modeling of hysteresis in piezoelectric actuator based on adaptive filter
CN113741183A (en) Industrial robot self-adaptive admittance control method based on damping ratio model
CN105929865A (en) Linear servo system mechanical resonance control method
CN103066602B (en) Microgrid mixed type mixed type power filter harmonic current prediction method based on back-direction optimum linear prediction theory
CN113110021B (en) A Method for Servo System Identification and Controller Design
CN115473467A (en) Flexible joint mechanical arm instruction filtering backstepping control method based on fuzzy observer
CN113093540B (en) A Design Method of Sliding Mode Disturbance Observer Based on Wavelet Threshold Denoising
CN103268403B (en) A kind of method for tracking target based on the strong trace information wave filter of volume
CN113625677A (en) Nonlinear system fault detection and estimation method and device based on adaptive iterative learning algorithm
CN110824922B (en) Smith prediction compensation method based on sixth-order B-spline wavelet neural network
CN109669348B (en) Industrial robot based on iterative feedback adjustment and trajectory tracking control method thereof
Xie et al. A robust two degree-of-freedom controller for systems with both model and measurement uncertainty
CN108123434B (en) Method for calculating slope of PV curve to obtain operating point of PV curve

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20221021

CF01 Termination of patent right due to non-payment of annual fee