WO2019024377A1 - 分数阶弧形微机电系统的自适应同步控制方法 - Google Patents

分数阶弧形微机电系统的自适应同步控制方法 Download PDF

Info

Publication number
WO2019024377A1
WO2019024377A1 PCT/CN2017/115740 CN2017115740W WO2019024377A1 WO 2019024377 A1 WO2019024377 A1 WO 2019024377A1 CN 2017115740 W CN2017115740 W CN 2017115740W WO 2019024377 A1 WO2019024377 A1 WO 2019024377A1
Authority
WO
WIPO (PCT)
Prior art keywords
fractional
order
adaptive
neural network
synchronization
Prior art date
Application number
PCT/CN2017/115740
Other languages
English (en)
French (fr)
Inventor
罗绍华
曹苏群
吴松励
侯志伟
陈中
Original Assignee
淮阴工学院
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 淮阴工学院 filed Critical 淮阴工学院
Publication of WO2019024377A1 publication Critical patent/WO2019024377A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04Adaptive 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
    • G05B13/042Adaptive 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
    • G05B13/045Adaptive 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 using a perturbation signal

Definitions

  • the invention relates to a fractional-order curved micro-electromechanical system, in particular to an adaptive synchronization control method of a fractional-order curved micro-electromechanical system.
  • Microelectromechanical systems are controllable, movable microelectromechanical devices whose outline dimensions are on the order of millimeters and whose constituent elements are on the order of micrometers. It is the inevitable result of people's constant pursuit of high-tech miniaturization since the advent of microelectronics technology.
  • Micro-electromechanical systems are micro-devices or systems that integrate micro-sensors, micro-actuators, micro-mechanical structures, micro-power micro-energy, signal processing and control circuits, high-performance electronic integrated devices, interfaces, and communications.
  • Micro-electromechanical system is a revolutionary new technology widely used in high-tech industries. It is a key technology related to the country's scientific and technological development, economic prosperity and national defense security.
  • fractional-order curved MEMS is more in line with the actual situation and more accurately reflects the engineering physics of the system. Influenced by uncertain factors, including interference, temperature, noise, measurement error, parameter estimation error and device aging, fractional-order curved MEMS has a high degree of uncertainty.
  • Time delay is an intrinsic property of the physical system. It is used to describe the hysteresis and the delay caused by inertia in the ideal state.
  • the fractional-order arc MEMS has the functions of unknown function, chaotic vibration and time delay under distributed electrostatic drive. Although the definition of chaotic synchronization is relaxed in logic, the synchronization control is complicated and the research is difficult.
  • the object of the present invention is to solve the problem of synchronous control of fractional-order curved micro-electromechanical system with unknown function, chaotic vibration and time delay under distributed electrostatic excitation, and the invention provides an adaptive of fractional-order curved micro-electromechanical system. Synchronous control method reduces the influence of uncertain factors on system stability and reliability, and realizes synchronous control of drive system and response system.
  • the present invention provides an adaptive synchronization control method for a fractional-order curved micro-electromechanical system, which is characterized in that it comprises the following steps:
  • Step S1 establishing a drive system and a response system model of the fractional-order arc MEMS based on the Euler-Bernoulli beam, and obtaining a synchronization error vector;
  • Step S2 constructing a Chebyshev neural network with an adaptive control law based on the error vector, constructing a virtual control input using a fractional order Liyanov function; using a Chebyshev neural network to estimate the unknown nonlinear function of the system, combined with fractional order
  • the adaptive law constitutes the actual control input, constructs an adaptive synchronization controller in the backstepping framework, and inputs the controller's output signal to the response system in the fractional-order arc MEMS.
  • the drive system model of the fractional-order arc MEMS based on the Euler-Bernoulli beam is:
  • ⁇ 1 denotes a fractional order
  • denotes a time delay
  • L denotes a length
  • A denotes a cross-sectional area
  • b denotes a width
  • C v denotes a viscous damping coefficient
  • d denotes a thickness
  • I y moment of inertia
  • density
  • ⁇ 0 harmonic load frequency
  • ⁇ a0 represents vacuum dielectric constant
  • V DC represents DC voltage
  • V AC AC voltage amplitude
  • the response system model is:
  • u(t) represents the control input of the response system.
  • the virtual control input process is constructed by using a fractional order Liyanov function
  • the virtual control is selected as:
  • the Chebyshev neural network is used to estimate the unknown nonlinear term of the system, and the process of constructing the actual control input by combining the backstepping, the Young's inequality and the fractional order adaptive law is:
  • the weight vector representing the Chebyshev neural network ⁇ 2 (e 1 , e 2 ) represents the basis function of the Chebyshev neural network, a 2 >0, Indicates an estimate of ⁇ 2 (t);
  • the adaptive synchronization controller is designed to:
  • the Chebyshev polynomial has an order of 3, and the basis function of the Chebyshev polynomial is
  • the invention achieves the beneficial effects that the adaptive synchronization control technology of a fractional-order curved micro-electromechanical system provided by the invention has an unknown function, chaotic vibration and
  • the fractional-order arc MEMS based on time delay is used as the object, and the mathematical model of the drive system and response system of fractional-order arc MEMS based on Euler-Bernoulli beam is established to realize the dynamic characteristics of the controlled system.
  • synchronization error vector use the Chebyshev neural network to arbitrarily Small errors approximate the characteristics of nonlinear functions, eliminating the requirement for accurate mathematical models and precise parameters of the system, using fractional-order Liyanov functions to solve virtual control inputs and actual control inputs and performing stability analysis in the framework of backstepping Construct an adaptive synchronization controller.
  • the invention realizes the synchronous control of the drive system and the response system to ensure the transient and steady state performance of the system, and reduces the uncertainty factors of the fractional-order arc MEMS (the existence of unknown function, parameter perturbation and time delay) for synchronous control The impact of performance.
  • Figure 1 is a schematic block diagram of the method of the present invention
  • Figure 2 is a schematic diagram of a fractional-order curved MEMS
  • Figure 3 is a phase diagram of a fractional-order curved microelectromechanical system with different excitation amplitudes
  • Figure 4 is a time sequence diagram of the state variables of the fractional-order arc MEMS under different excitation amplitudes
  • FIG. 5 is a time sequence diagram of system states x 1 , y 1 under different excitation amplitudes in an embodiment of the present invention
  • FIG. 6 is a time sequence diagram of system states x 2 , y 2 under different excitation amplitudes in an embodiment of the present invention
  • Figure 9 is a graph showing the control input of the system at different excitation amplitudes in an embodiment of the present invention.
  • Chebyshev polynomial is defined as:
  • T 1 (X) is generally defined as X, 2X, 2X-1 or 2X+1.
  • the Chebyshev neural network has the ability to approximate any nonlinear continuous function with a arbitrarily small precision error on a compact set.
  • X [x 1 ,..., x m ] T ⁇ R m , define the Chebyshev polynomial Enhanced mode is
  • ⁇ (X) [1,T 1 (x 1 ),...,T n (x 1 ),...,T 1 (x m ),...,T n (x m )]
  • n denotes an order
  • ⁇ (X) denotes a base of a Chebyshev polynomial function.
  • ⁇ (t) represents the weight vector
  • n represents an integer and has n-1 ⁇ ⁇ ⁇ n, and ⁇ represents a fractional order.
  • Theorem 1 Defining fractional Lyapunov functions It is a continuous differentiable function and is relative to x(t) local Lipschitz. If there is a K class function
  • the Mittag-Leffler function is defined as follows:
  • the adaptive synchronization control method of the fractional-order curved micro-electromechanical system of the invention is based on a fractional-order curved micro-electromechanical system with unknown functions, chaotic vibration and time delay under distributed electrostatic excitation, as shown in Fig. 1. Show, including the following steps:
  • step S1 a drive system and a response system model of a fractional-order arc MEMS based on Euler-Bernoulli beam are established, and a synchronization error vector is obtained.
  • ⁇ 1 denotes a fractional order
  • denotes a time delay
  • L denotes a length
  • A denotes a cross-sectional area
  • b denotes a width
  • C v denotes a viscous damping coefficient
  • d denotes a thickness
  • I y moment of inertia
  • density
  • ⁇ 0 harmonic load frequency
  • ⁇ a0 represents vacuum dielectric constant
  • V DC represents DC voltage
  • V AC AC voltage amplitude.
  • the Runge-Kutta algorithm is used to solve the solution, and the solution time is more than 1000 seconds.
  • Fig. 3 is a phase diagram of different excitation amplitudes (R values)
  • Fig. 4 is a time sequence diagram of state variables of fractional arc MEMS under different excitation amplitudes
  • Fig. 3 and Fig. 4 suggest fractional arc micros
  • the electromechanical system exhibits chaotic oscillation under different amplitude excitations.
  • u(t) represents the control input of the response system and is also the response system controller to be designed (synchronization requirement).
  • Step S2 constructing a Chebyshev neural network with an adaptive control law based on the error vector, constructing a virtual control input using a fractional order Liyanov function; using a Chebyshev neural network to estimate the unknown nonlinear function of the system, combined with fractional order
  • the adaptive law constitutes the actual control input, constructs an adaptive synchronization controller in the backstepping framework, and inputs the controller's output signal to the response system in the fractional-order arc MEMS.
  • the design of the adaptive synchronization controller includes the following process:
  • Step S21 constructing a virtual control input by using a fractional order Liyanov function.
  • the virtual control is selected as:
  • Step S22 using the Chebyshev neural network to estimate the unknown nonlinear term of the system, canceling the requirements of the accurate mathematical model and the precise parameters of the system, and combining the backstepping, Young's inequality and fractional order adaptive law to form the actual control input.
  • the weight vector representing the Chebyshev neural network ⁇ 2 (e 1 , e 2 ) represents the basis function of the Chebyshev neural network, a 2 >0, Represents an estimate of ⁇ 2 (t).
  • f 2 (e 1 , e 2 ) contains a variable function term and a cosine function term of more than one power, and can be regarded as a nonlinear function.
  • the nonlinear function f 2 (e 1 , e 2 ) has very complex nonlinear characteristics. Accurate measurement of system parameters is very difficult due to external disturbances, manufacturing defects, and modeling errors. At the same time, fractional-order curved MEMS are very sensitive to system parameter perturbation, and some range of parameter values can cause chaotic oscillations in fractional-order arc MEMS. In order to solve the above problems and facilitate controller design, Chebyshev neural network is used to approximate nonlinear functions.
  • the upper limit of the error is estimated for the Chebyshev neural network.
  • the adaptive synchronization controller is designed to:
  • ⁇ 2 min(2k 1 , 2k 21 , c 2 ) and Indicates a normal number.
  • Theorem 2 For fractional-order arc MEMS with distributed function, unknown function, chaotic vibration and time delay, if fractional order law is designed as equation (14) and controller is designed as equation (13) Then, (S 1 (t), S 2 (t)) is globally asymptotically stable at the equilibrium point position, and the drive and response systems of the fractional-order arc-shaped micro-electric system resonator are synchronized.
  • V(s) is the V(t) Laplace transform term.
  • Equation (18) is transformed into Equation (18).
  • the initial value of the drive system x 1 (0), x 2 (0) is equal to zero, and the initial values y 1 (0), y 2 (0) of the response system are set to 0.01 and 0.02.
  • Figures 5 and 6 show the state variables x 1 , x 2 , y 1 , y 2 trajectories at different excitation amplitudes.
  • Figures 7 and 8 depict the synchronization error e 1 , e 2 trajectories at different excitation amplitudes.
  • Figure 9 shows the control inputs for fractional-order arc MEMS with different excitation amplitudes. It can be seen that the parameter perturbation has no effect on the stable synchronization of the fractional-order arc MEMS.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
  • Other Investigation Or Analysis Of Materials By Electrical Means (AREA)

Abstract

一种分数阶弧形微机电系统的自适应同步控制方法,包括以下步骤:步骤S1,建立基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统与响应系统模型,得到同步误差向量;步骤S2,根据误差向量构建带有自适应控制律的切比雪夫神经网络,利用分数阶李亚谱诺夫函数构建虚拟控制输入;利用切比雪夫神经网络估计系统未知非线性函数,结合分数阶自适应律构成实际控制输入,在backstepping的框架中构造自适应同步控制器,并将控制器的输出信号输入到分数阶弧形微机电系统中的响应系统。该方法保证系统瞬态和稳态性能的驱动系统与响应系统的同步控制,降低分数阶弧形微机电系统不确定性因素对同步控制性能的影响。

Description

分数阶弧形微机电系统的自适应同步控制方法 技术领域
本发明涉及分数阶弧形微机电系统,具体涉及一种分数阶弧形微机电系统的自适应同步控制方法。
背景技术
微型机电系统是指那些外形轮廓尺寸在毫米量级以下,构成元件是微米量级的可控制、可运动的微型机电装置。它是自微电子技术问世以来,人们不断追求高新技术微型化的必然结果。微型机电系统是集微型传感器、微型执行器、微型机械结构、微型电源微型能源、信号处理和控制电路、高性能电子集成器件、接口、通信等于一体的微型器件或系统。微型机电系统是一项革命性的新技术,广泛应用于高新技术产业,是一项关系到国家的科技发展、经济繁荣和国防安全的关键技术。
实际控制系统的分析与设计需要简单、准确的数学模型,而分数阶微积分理论能实现对被控系统中动力学特性精确的描述。与整数阶弧形微机电系统相比,分数阶弧形微机电系统更符合实际情况,更能准确反应系统的工程物理现象。受到不确定因素的影响,包括干扰、温度、噪声、测量误差、参数估计误差与器件老化等,分数阶弧形微机电系统具有高度的不确定性。时延是物理系统的固有特性,用来描述在理想状态下传送过程中的滞后现象和惯性作用所导致的时延现象。分数阶弧形微机电系统在分布式静电驱动下具有未知函数,混沌振动和时延等特征,虽在逻辑上放宽了混沌同步的定义,但其同步控制复杂,研究的难度大。
发明内容
本发明的目的在于解决分数阶弧形微机电系统在分布式静电激励下具有未知函数,混沌振动和时延等特征的同步控制问题,本发明提供一种分数阶弧形微机电系统的自适应同步控制方法,降低不确定因素对系统稳定性与可靠性的影响,实现驱动系统与响应系统的同步控制。
为解决上述技术问题,本发明提供了分数阶弧形微机电系统的自适应同步控制方法,其特征是,包括以下步骤:
步骤S1,建立基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统与响应系统模型,得到同步误差向量;
步骤S2,根据误差向量构建带有自适应控制律的切比雪夫神经网络,利用分数阶李亚谱诺夫函数构建虚拟控制输入;利用切比雪夫神经网络估计系统未知非线性函数,结合分数阶自适应律构成实际控制输入,在backstepping的框架中构造自适应同步控制器,并将控制器的输出信号输入到分数阶弧形微机电系统中的响应系统。
优选的,基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统模型为:
Figure PCTCN2017115740-appb-000001
其中各变量定义如下:
Figure PCTCN2017115740-appb-000002
α1表示分数阶,τ表示时延,L表示长度,A表示横截面积,b表示宽度,Cv 表示粘滞阻尼系数,d表示厚度,
Figure PCTCN2017115740-appb-000003
表示杨氏模量,Iy表示转动惯量,ρ表示密度,Ω0表示谐波负载频率,εa0表示真空介电常数,VDC表示直流电压,VAC表示交流电压幅值;
响应系统模型为:
Figure PCTCN2017115740-appb-000004
其中u(t)表示响应系统的控制输入。
优选的,定义同步误差为ei=yi-xi,i=1,2;结合式(1)(2),得出同步误差向量为:
Figure PCTCN2017115740-appb-000005
优选的,利用分数阶李亚谱诺夫函数构建虚拟控制输入过程为;
定义S1(t)=e1(t),选取分数阶李亚谱诺夫函数:
Figure PCTCN2017115740-appb-000006
相应的其分数阶导数为:
Figure PCTCN2017115740-appb-000007
其中S2(t)=e2(t)-αv2(t),αv2(t)表示虚拟控制输入;
根据式(5),选取虚拟控制为:
αv2(t)=-k1S1(t)                 (6)
其中k1>0;
把式(6)代入式(5)得到
Figure PCTCN2017115740-appb-000008
优选的,利用切比雪夫神经网络估计系统未知非线性项,结合backstepping、杨氏不等式和分数阶自适应律构成实际控制输入的过程为:
基于杨氏不等式,将两个参数变量
Figure PCTCN2017115740-appb-000009
和ξ2(e1,e2)代入得到
Figure PCTCN2017115740-appb-000010
其中
Figure PCTCN2017115740-appb-000011
Figure PCTCN2017115740-appb-000012
表示切比雪夫神经网络的权向量,ξ2(e1,e2)表示切比雪夫神经网络的基函数,a2>0,
Figure PCTCN2017115740-appb-000013
Figure PCTCN2017115740-appb-000014
表示ζ2(t)的估计值;
选择分数阶李雅普诺夫函数
Figure PCTCN2017115740-appb-000015
其中μ2>0;
其相应的分阶导数为:
Figure PCTCN2017115740-appb-000016
定义非线性函数f2(e1,e2)为:
Figure PCTCN2017115740-appb-000017
采用切比雪夫神经网络来逼近非线性函数
Figure PCTCN2017115740-appb-000018
由于常数的分数阶导数等于零,则
Figure PCTCN2017115740-appb-000019
把切比雪夫神经网络和式(8)代入式(10),可以得到:
Figure PCTCN2017115740-appb-000020
其中
Figure PCTCN2017115740-appb-000021
Figure PCTCN2017115740-appb-000022
为切比雪夫神经网络估计误差的上限;
根据式(12),自适应同步控制器设计为:
Figure PCTCN2017115740-appb-000023
其中k21>0,k22>0;
根据式(12)和式(13),分数阶自适应律设计为:
Figure PCTCN2017115740-appb-000024
其中c2>0。
优选的,根据分数阶弧形微机电系统的同步误差向量式(3)可知,切比雪夫多项式的阶数为3,那么切比雪夫多项式的基函数为
Figure PCTCN2017115740-appb-000025
与现有技术相比,本发明所达到的有益效果是:本发明提供的一种分数阶弧形微机电系统的自适应同步控制技术,以在分布式静电激励下具有未知函数,混沌振动和时延等特征的分数阶弧形微机电系统为对象,建立基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统与响应系统数学模型,实现对被控系统中动力学特性精确的描述,定义同步误差向量,利用切比雪夫神经网络以任意 小的误差逼近非线性函数的特性,取消了对系统精确数学模型与精准参数的要求,利用分数阶李亚谱诺夫函数求解虚拟控制输入与实际控制输入并进行稳定性分析,在backstepping的框架中构造自适应同步控制器。本发明实现了保证系统瞬态和稳态性能的驱动系统与响应系统的同步控制,降低分数阶弧形微机电系统不确定性因素(存在未知函数、参数摄动和时延时)对同步控制性能的影响。
附图说明
图1是本发明方法的原理框图;
图2是分数阶弧形微机电系统的原理图;
图3是不同激励幅值下分数阶弧形微机电系统的相图;
图4是不同激励幅值下分数阶弧形微机电系统状态变量的时间序图;
图5是本发明实施例中不同激励幅值下系统状态x1,y1的时间序图;
图6是本发明实施例中不同激励幅值下系统状态x2,y2的时间序图;
图7是本发明实施例中不同激励幅值下同步误差e1曲线图;
图8是本发明实施例中不同激励幅值下同步误差e2曲线图;
图9是本发明实施例中不同激励幅值下系统的控制输入曲线图。
具体实施方式
下面结合附图对本发明作进一步描述。以下实施例仅用于更加清楚地说明本发明的技术方案,而不能以此来限制本发明的保护范围。
下面先对下文中出现的理论做详细的描述:
(1)切比雪夫神经网络
通过在两项递推公式中求解不等式,切比雪夫多项式定义为:
Ti+1(X)=2XTi(X)-Ti-1(X),T0(X)=1
其中X∈R,T1(X)通常被定义为X,2X,2X-1或2X+1。
切比雪夫神经网络具有在一个紧凑集上以任意小的精度误差逼近任意非线性连续函数的能力,对于X=[x1,...,xm]T∈Rm,定义切比雪夫多项式的增强模式为
ξ(X)=[1,T1(x1),...,Tn(x1),...,T1(xm),...,Tn(xm)]
其中Ti(xj),i=1,...,n,j=1,...,m,表示切比雪夫多项式,n表示阶数,ξ(X)表示切比雪夫多项式的基函数。
未知非线性项fcnn(X)的估计值
Figure PCTCN2017115740-appb-000026
可以由切比雪夫神经网络求取,则
Figure PCTCN2017115740-appb-000027
其中φ(t)表示权向量。
对于上式,存在
Figure PCTCN2017115740-appb-000028
其中ε>0,Ωφ和DX分别表示φ(t)和X的紧凑集。让最优参数φ*等于
Figure PCTCN2017115740-appb-000029
出于分析目的,
Figure PCTCN2017115740-appb-000030
存在。
(2)分数阶理论定义
Caputo分数阶导数定义为
Figure PCTCN2017115740-appb-000031
其中
Figure PCTCN2017115740-appb-000032
表示欧拉函数,n表示整数并有n-1≤α≤n,α表示分数阶。
求Caputo分数阶导数的分拉普拉斯变换,则
Figure PCTCN2017115740-appb-000033
定理1:定义分数阶李雅普诺夫函数
Figure PCTCN2017115740-appb-000034
为连续可微函数且相 对于x(t)局部李普希茨。如果存在K类函数则
Figure PCTCN2017115740-appb-000035
其中t>0,x(0)渐近稳定。
Mittag-Leffler函数定义如下:
Figure PCTCN2017115740-appb-000036
其中ν>0,δ>0,γ表示复数。
对上式Mittag-Leffler函数进行拉普拉斯变换得到
Figure PCTCN2017115740-appb-000037
对于0<ν<1和
Figure PCTCN2017115740-appb-000038
存在
Figure PCTCN2017115740-appb-000039
当|γ|→∞,σ1≤|arg(γ)|≤π。
对于0<ν<2和
Figure PCTCN2017115740-appb-000040
下列关系存在
Figure PCTCN2017115740-appb-000041
其中B>0,σ2≤|arg(γ)|≤π和|γ|≥0。
本发明的分数阶弧形微机电系统的自适应同步控制方法,以在分布式静电激励下具有未知函数,混沌振动和时延等特征的分数阶弧形微机电系统为对象,如图1所示,包括以下步骤:
步骤S1,建立基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统与响应系统模型,得到同步误差向量。
分数阶弧形微机电系统的原理图如图2所示。基于欧拉-伯努利梁的分数阶 弧形微机电系统的动力学模型为:
Figure PCTCN2017115740-appb-000042
其中各变量定义如下:
Figure PCTCN2017115740-appb-000043
α1表示分数阶,τ表示时延,L表示长度,A表示横截面积,b表示宽度,Cv表示粘滞阻尼系数,d表示厚度,
Figure PCTCN2017115740-appb-000044
表示杨氏模量,Iy表示转动惯量,ρ表示密度,Ω0表示谐波负载频率,εa0表示真空介电常数,VDC表示直流电压,VAC表示交流电压幅值。
此系统非线性行为分析:定义系统参数αm=7.993,β=119.9883,h=0.3,μ=0.1,α1=0.98,τ=0.1和ω0=0.4706,进而对分数阶弧形微机电系统进行非线性动力学分析。为了求解分数阶导数方程,应用龙格-库塔算法进行求解,其求解时间大于1000秒。图3是不同激励幅值(R值)下的相图,图4是不同激励幅值下分数阶弧形微机电系统状态变量的时间序图,图3和图4提示了分数阶弧形微机电系统在不同的幅值激励下出现了混沌振荡现象。
考虑分数阶弧形微机电系统的驱动和响应系统。不失一般性,驱动分数阶弧形微机电系统的数学模型如式(1)所示,响应分数阶弧形微机电系统的数学模型可写为:
Figure PCTCN2017115740-appb-000045
其中u(t)表示响应系统的控制输入,也是待设计的响应系统控制器(同步要求)。
建立基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统与响应系统模型,实现对被控系统中动力学特性精确的描述,并定义同步误差为ei=yi-xi,i=1,2。
结合式(1)(2),得出同步误差向量为:
Figure PCTCN2017115740-appb-000046
本发明控制目的为:通过设计控制器u(t)使得驱动和响应的分数阶弧形微机电系统能够实现全局同步,即同步误差渐近趋于0,即
Figure PCTCN2017115740-appb-000047
其中,e(t)=[e1,e2]T
步骤S2,根据误差向量构建带有自适应控制律的切比雪夫神经网络,利用分数阶李亚谱诺夫函数构建虚拟控制输入;利用切比雪夫神经网络估计系统未知非线性函数,结合分数阶自适应律构成实际控制输入,在backstepping的框架中构造自适应同步控制器,并将控制器的输出信号输入到分数阶弧形微机电系统中的响应系统。
自适应同步控制器的设计包括以下过程:
步骤S21:利用分数阶李亚谱诺夫函数构建虚拟控制输入。
定义S1(t)=e1(t),e1(t)是误差向量,选取分数阶李亚谱诺夫函数(李亚谱诺夫函数是公知的求解控制器的通用方法)
Figure PCTCN2017115740-appb-000048
相应的其分数阶导数为:
Figure PCTCN2017115740-appb-000049
其中S2(t)=e2(t)-αv2(t),αv2(t)表示虚拟控制输入。
根据式(5),选取虚拟控制为:
αv2(t)=-k1S1(t)              (6)
其中k1>0。
把式(6)代入式(5)得到
Figure PCTCN2017115740-appb-000050
步骤S22,利用切比雪夫神经网络估计系统未知非线性项,取消了对系统精确数学模型与精准参数的要求,结合backstepping、杨氏不等式和分数阶自适应律构成实际控制输入。
根据分数阶弧形微机电系统的同步误差向量式(3)可知,切比雪夫多项式的阶数为3,那么切比雪夫多项式的基函数为
Figure PCTCN2017115740-appb-000051
为了减轻在线计算的负担,采取减少切比雪夫神经网络权向量的数量。基于杨氏不等式,将两个参数变量
Figure PCTCN2017115740-appb-000052
和ξ2(e1,e2)代入得到
Figure PCTCN2017115740-appb-000053
其中
Figure PCTCN2017115740-appb-000054
Figure PCTCN2017115740-appb-000055
表示切比雪夫神经网络的权向量,ξ2(e1,e2)表示切比 雪夫神经网络的基函数,a2>0,
Figure PCTCN2017115740-appb-000056
Figure PCTCN2017115740-appb-000057
表示ζ2(t)的估计值。
选择分数阶李雅普诺夫函数
Figure PCTCN2017115740-appb-000058
其中μ2>0。
其相应的分阶导数为:
Figure PCTCN2017115740-appb-000059
定义非线性函数f2(e1,e2)为:
Figure PCTCN2017115740-appb-000060
f2(e1,e2)包涵了1次方以上的变量函数项及余弦函数项,即可认定为非线性函数。非线性函数f2(e1,e2)具有非常复杂的非线性特征。由于受到外界扰动、制造缺陷和建模误差的影响,对系统参数进行精确测量变得非常困难。同时,分数阶弧形微机电系统对系统参数摄动非常敏感,某些范围内的参数值可导致分数阶弧形微机电系统出现混沌振荡。为了解决上述问题和便于控制器设计,采用切比雪夫神经网络来逼近非线性函数
Figure PCTCN2017115740-appb-000061
由于常数的分数阶导数等于零,则
Figure PCTCN2017115740-appb-000062
把切比雪夫神经网络和式(8)代入式(10),可以得到:
Figure PCTCN2017115740-appb-000063
其中
Figure PCTCN2017115740-appb-000064
Figure PCTCN2017115740-appb-000065
为切比雪夫神经网络估计误差的上限。
根据式(12),自适应同步控制器设计为:
Figure PCTCN2017115740-appb-000066
其中k21>0,k22>0。
根据式(12)和式(13),分数阶自适应律设计为:
Figure PCTCN2017115740-appb-000067
其中c2>0。
从式(13),式(14)和式(12)可知,存在
Figure PCTCN2017115740-appb-000068
其中κ2=min(2k1,2k21,c2)和
Figure PCTCN2017115740-appb-000069
表示正常数。
系统稳定性证明:
定理2:针对分数阶弧形微机电系统在分布式静电驱动下具有未知函数,混沌振动和时延等特征,假如分数阶自适应律设计为式(14)和控制器设计为式(13),那么(S1(t),S2(t))在平衡点位置全局渐近稳定,分数阶弧形微电系统谐振器的驱动和响应系统实现同步。
证明:定义分数阶李亚谱函数
Figure PCTCN2017115740-appb-000070
对分数阶李亚谱函数进行求导
Figure PCTCN2017115740-appb-000071
对式(17)进行拉普拉斯变换
Figure PCTCN2017115740-appb-000072
其中V(s)是V(t)拉普拉斯变换项。
根据定理1,式(18)变换为
Figure PCTCN2017115740-appb-000073
存在正定常数B满足关系式
Figure PCTCN2017115740-appb-000074
基于式(20),得到
Figure PCTCN2017115740-appb-000075
对任意θ>0,存在正定常数t1>t,则
Figure PCTCN2017115740-appb-000076
存在
Figure PCTCN2017115740-appb-000077
存在正定常数t2>t,得到
Figure PCTCN2017115740-appb-000078
调整参数使其满足关系式
Figure PCTCN2017115740-appb-000079
同时借助于式(19),式(22)和式(24),得到
|V(t)|≤θ
证明完毕。
为了验证本发明方法的控制效果,进行以下仿真试验:
控制器参数选取为k1=3,k21=3,k22=0.03,a2=1,c2=0.8,μ2=0.8。驱动系统的初始值x1(0),x2(0)等于零,响应系统的初始值y1(0),y2(0)设置为0.01和0.02。
图5和图6展示了在不同激励幅值下的状态变量x1,x2,y1,y2轨迹。图7和图8描述了在不同激励幅值下的同步误差e1,e2轨迹。在所述控制器未被应用前,分数阶弧形微机电系统的驱动系统和响应系统产生混沌现象,同步没有实现。当在第20秒时应用所述控制器,自适应同步方案保证了系统状态稳定在即定轨迹上。很明显,同步误差趋于零,所述控制方案实现了驱动系统和响应系统的同步。
尽管分数阶弧形微机电系统工作时受到未知函数,时延和参数扰动等的影响,与图4进行对比,所提控制方案很好的解决这些问题。图9展示了分数阶弧形微机电系统在不同激励幅值下的控制输入。可以看出,参数扰动对分数阶弧形微机电系统的稳定同步并没有影响。
以上所述仅是本发明的优选实施方式,应当指出,对于本技术领域的普通技术人员来说,在不脱离本发明技术原理的前提下,还可以做出若干改进和变型,这些改进和变型也应视为本发明的保护范围。

Claims (6)

  1. 分数阶弧形微机电系统的自适应同步控制方法,其特征是,包括以下步骤:
    步骤S1,建立基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统与响应系统模型,得到同步误差向量;
    步骤S2,根据误差向量构建带有自适应控制律的切比雪夫神经网络,利用分数阶李亚谱诺夫函数构建虚拟控制输入;利用切比雪夫神经网络估计系统未知非线性函数,结合分数阶自适应律构成实际控制输入,在backstepping的框架中构造自适应同步控制器,并将控制器的输出信号输入到分数阶弧形微机电系统中的响应系统。
  2. 根据权利要求1所述的分数阶弧形微机电系统的自适应同步控制方法,其特征是,基于欧拉-伯努利梁的分数阶弧形微机电系统的驱动系统模型为:
    Figure PCTCN2017115740-appb-100001
    其中各变量定义如下:
    Figure PCTCN2017115740-appb-100002
    Figure PCTCN2017115740-appb-100003
    α1表示分数阶,τ表示时延,L表示长度,A表示横截面积,b表示宽度,Cv表示粘滞阻尼系数,d表示厚度,
    Figure PCTCN2017115740-appb-100004
    表示杨氏模量,Iy表示转动惯量,ρ表示密度,Ω0表示谐波负载频率,εa0表示真空介电常数,VDC表示直流电压,VAC表示交流电压幅值;
    响应系统模型为:
    Figure PCTCN2017115740-appb-100005
    其中u(t)表示响应系统的控制输入。
  3. 根据权利要求1所述的分数阶弧形微机电系统的自适应同步控制方法,其特征是,定义同步误差为ei=yi-xi,i=1,2;结合式(1)(2),得出同步误差向量为:
    Figure PCTCN2017115740-appb-100006
  4. 根据权利要求1所述的分数阶弧形微机电系统的自适应同步控制方法,其特征是,利用分数阶李亚谱诺夫函数构建虚拟控制输入过程为;
    定义S1(t)=e1(t),选取分数阶李亚谱诺夫函数:
    Figure PCTCN2017115740-appb-100007
    相应的其分数阶导数为:
    Figure PCTCN2017115740-appb-100008
    其中S2(t)=e2(t)-αv2(t),αv2(t)表示虚拟控制输入;
    根据式(5),选取虚拟控制为:
    αv2(t)=-k1S1(t)  (6)
    其中k1>0;
    把式(6)代入式(5)得到
    Figure PCTCN2017115740-appb-100009
  5. 根据权利要求1所述的分数阶弧形微机电系统的自适应同步控制方法,其特征是,设计自适应同步控制器的具体过程为:
    基于杨氏不等式,将两个参数变量
    Figure PCTCN2017115740-appb-100010
    和ξ2(e1,e2)代入得到
    Figure PCTCN2017115740-appb-100011
    其中
    Figure PCTCN2017115740-appb-100012
    表示切比雪夫神经网络的权向量,ξ2(e1,e2)表示切比雪夫神经网络的基函数,a2>0,
    Figure PCTCN2017115740-appb-100013
    表示ζ2(t)的估计值;
    选择分数阶李雅普诺夫函数
    Figure PCTCN2017115740-appb-100014
    其中μ2>0;
    其相应的分阶导数为:
    Figure PCTCN2017115740-appb-100015
    定义非线性函数f2(e1,e2)为:
    Figure PCTCN2017115740-appb-100016
    采用切比雪夫神经网络来逼近非线性函数
    Figure PCTCN2017115740-appb-100017
    由于常数的分数阶导数等于零,则
    Figure PCTCN2017115740-appb-100018
    把切比雪夫神经网络和式(8)代入式(10),可以得到:
    Figure PCTCN2017115740-appb-100019
    其中
    Figure PCTCN2017115740-appb-100020
    为切比雪夫神经网络估计误差的上限;
    根据式(12),自适应同步控制器设计为:
    Figure PCTCN2017115740-appb-100021
    其中k21>0,k22>0;
    根据式(12)和式(13),分数阶自适应律设计为:
    Figure PCTCN2017115740-appb-100022
    其中c2>0。
  6. 根据权利要求1所述的分数阶弧形微机电系统的自适应同步控制方法,其特征是,根据分数阶弧形微机电系统的同步误差向量式(3)可知,切比雪夫多项式的阶数为3,那么切比雪夫多项式的基函数为
    Figure PCTCN2017115740-appb-100023
PCT/CN2017/115740 2017-08-03 2017-12-12 分数阶弧形微机电系统的自适应同步控制方法 WO2019024377A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710655332.7A CN107479377B (zh) 2017-08-03 2017-08-03 分数阶弧形微机电系统的自适应同步控制方法
CN201710655332.7 2017-08-03

Publications (1)

Publication Number Publication Date
WO2019024377A1 true WO2019024377A1 (zh) 2019-02-07

Family

ID=60597601

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/115740 WO2019024377A1 (zh) 2017-08-03 2017-12-12 分数阶弧形微机电系统的自适应同步控制方法

Country Status (2)

Country Link
CN (1) CN107479377B (zh)
WO (1) WO2019024377A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109807902A (zh) * 2019-04-08 2019-05-28 青岛大学 一种基于反步法的双机械臂力/位模糊混合控制方法
CN110134011A (zh) * 2019-04-23 2019-08-16 浙江工业大学 一种倒立摆自适应迭代学习反演控制方法
CN113064347A (zh) * 2021-03-15 2021-07-02 贵州大学 考虑非对称输入与输出约束的pmsm混沌系统自适应控制方法
CN113885463A (zh) * 2021-10-26 2022-01-04 盐城工学院 一种畜禽生物菌肥好氧发酵罐工艺参数调控系统和方法
GB2598416A (en) * 2019-12-16 2022-03-02 Chongqing Aerospace Polytechnic Optimal sychronization control method for coupled fractional-order chaotic electromechanical device
CN114839880A (zh) * 2022-06-02 2022-08-02 淮阴工学院 一种基于柔性关节机械臂的自适应控制方法
CN116400598A (zh) * 2023-04-07 2023-07-07 盐城工学院 一种惯性cg神经网络的同步控制方法

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108614419B (zh) * 2018-03-28 2020-12-08 贵州大学 一种弧形微机电系统的自适应神经网络控制方法
CN109634116B (zh) * 2018-09-04 2022-03-15 贵州大学 一种分数阶机械式离心调速器系统的加速自适应稳定方法
CN109188914B (zh) * 2018-10-26 2021-04-27 黑龙江大学 一种n阶混合非线性系统的协同控制方法及控制系统
CN109613826B (zh) * 2018-12-17 2021-07-27 重庆航天职业技术学院 一种分数阶拱形mems谐振器的反振荡自适应控制方法
CN109991852B (zh) * 2019-04-19 2022-02-22 贵州大学 具有磁滞特性的分数阶静电驱动微机电系统控制方法
CN109946978B (zh) * 2019-04-25 2022-02-08 广东省智能机器人研究院 一种考虑延时环节的伺服系统分数阶模型辨识方法
CN112965383B (zh) * 2021-02-09 2022-08-09 重庆航天职业技术学院 单向耦合分数阶自持机电地震仪系统的自适应神经网络最优定时同步控制方法
CN113095497B (zh) * 2021-05-06 2023-04-07 安徽大学 一种分数阶四元忆阻神经网络的有限时间同步方法、装置
CN113219836B (zh) * 2021-05-19 2022-04-08 安徽大学 一种分数阶复值忆阻神经网络的投影同步方法及其应用
CN113777920B (zh) * 2021-08-19 2023-09-15 中国人民解放军海军航空大学 基于rbf-nn及观测器的分数阶混沌同步控制方法
CN115508665A (zh) * 2022-09-23 2022-12-23 淮阴工学院 一种针对rlc分数线性电路的故障检测方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6560493B1 (en) * 1999-02-26 2003-05-06 Massachusetts Institute Of Technology Architecture for distributed control of actuator and sensor arrays
CN103645646A (zh) * 2013-12-04 2014-03-19 冶金自动化研究设计院 内含微小电感或电容的复杂电路系统建模与控制方法
CN105449699A (zh) * 2016-01-11 2016-03-30 东北电力大学 双馈感应风电机组非线性分数阶自抗扰阻尼控制方法
CN106325073A (zh) * 2016-10-31 2017-01-11 华中科技大学 基于分数阶的伺服系统位置ip控制器无模型自校正方法
CN106647277A (zh) * 2017-01-06 2017-05-10 淮阴工学院 弧形微型机电混沌系统的自适应动态面控制方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1924736A (zh) * 2006-09-21 2007-03-07 上海大学 基于前庭动眼反射的航空侦察平台稳定控制方法
CN100547509C (zh) * 2008-04-02 2009-10-07 华南理工大学 基于加速度传感器的挠性悬臂板振动控制装置与控制方法
WO2013182901A1 (en) * 2012-06-07 2013-12-12 Actiwave Ab Non-linear control of loudspeakers
US8972315B2 (en) * 2012-10-25 2015-03-03 Brain Corporation Apparatus and methods for activity-based plasticity in a spiking neuron network
CN105204343B (zh) * 2015-10-13 2018-05-15 淮阴工学院 具有输出约束和死区输入的纳米机电系统backstepping控制方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6560493B1 (en) * 1999-02-26 2003-05-06 Massachusetts Institute Of Technology Architecture for distributed control of actuator and sensor arrays
CN103645646A (zh) * 2013-12-04 2014-03-19 冶金自动化研究设计院 内含微小电感或电容的复杂电路系统建模与控制方法
CN105449699A (zh) * 2016-01-11 2016-03-30 东北电力大学 双馈感应风电机组非线性分数阶自抗扰阻尼控制方法
CN106325073A (zh) * 2016-10-31 2017-01-11 华中科技大学 基于分数阶的伺服系统位置ip控制器无模型自校正方法
CN106647277A (zh) * 2017-01-06 2017-05-10 淮阴工学院 弧形微型机电混沌系统的自适应动态面控制方法

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109807902A (zh) * 2019-04-08 2019-05-28 青岛大学 一种基于反步法的双机械臂力/位模糊混合控制方法
CN110134011A (zh) * 2019-04-23 2019-08-16 浙江工业大学 一种倒立摆自适应迭代学习反演控制方法
CN110134011B (zh) * 2019-04-23 2022-01-11 浙江工业大学 一种倒立摆自适应迭代学习反演控制方法
GB2598416A (en) * 2019-12-16 2022-03-02 Chongqing Aerospace Polytechnic Optimal sychronization control method for coupled fractional-order chaotic electromechanical device
CN113064347A (zh) * 2021-03-15 2021-07-02 贵州大学 考虑非对称输入与输出约束的pmsm混沌系统自适应控制方法
CN113885463A (zh) * 2021-10-26 2022-01-04 盐城工学院 一种畜禽生物菌肥好氧发酵罐工艺参数调控系统和方法
CN113885463B (zh) * 2021-10-26 2023-09-01 盐城工学院 一种畜禽生物菌肥好氧发酵罐工艺参数调控系统和方法
CN114839880A (zh) * 2022-06-02 2022-08-02 淮阴工学院 一种基于柔性关节机械臂的自适应控制方法
CN114839880B (zh) * 2022-06-02 2024-04-19 淮阴工学院 一种基于柔性关节机械臂的自适应控制方法
CN116400598A (zh) * 2023-04-07 2023-07-07 盐城工学院 一种惯性cg神经网络的同步控制方法
CN116400598B (zh) * 2023-04-07 2023-09-26 盐城工学院 一种惯性cg神经网络的同步控制方法

Also Published As

Publication number Publication date
CN107479377B (zh) 2020-06-12
CN107479377A (zh) 2017-12-15

Similar Documents

Publication Publication Date Title
WO2019024377A1 (zh) 分数阶弧形微机电系统的自适应同步控制方法
CN108614419B (zh) 一种弧形微机电系统的自适应神经网络控制方法
CN109613826B (zh) 一种分数阶拱形mems谐振器的反振荡自适应控制方法
CN109991852B (zh) 具有磁滞特性的分数阶静电驱动微机电系统控制方法
CN106647277B (zh) 弧形微型机电混沌系统的自适应动态面控制方法
Afshari et al. Global sliding mode control via linear matrix inequality approach for uncertain chaotic systems with input nonlinearities and multiple delays
Agrawal et al. An analytical formulation for phase noise in MEMS oscillators
US10768585B2 (en) System and method for data-driven control with partially unknown feedback
Fei et al. Adaptive nonsingular terminal sliding mode control of MEMS gyroscope based on backstepping design
Han et al. Design considerations on large amplitude vibration of a doubly clamped microresonator with two symmetrically located electrodes
Trapero et al. A fast on-line frequency estimator of lightly damped vibrations in flexible structures
Shitole et al. Sliding DFT-based vibration mode estimator for single-link flexible manipulator
Phani et al. Modal coupling in micromechanical vibratory rate gyroscopes
Tuwa et al. Suppression of the noise-induced effects in an electrostatic micro-plate using an adaptive back-stepping sliding mode control
Carvajal-Rubio et al. On the discretization of a class of homogeneous differentiators
Liu et al. Bounded adaptive output feedback tracking control for flexible-joint robot manipulators
Petronijević et al. Robust cascade control of electrical drives using discrete-time chattering-free sliding mode controllers with output saturation
Zhou et al. Speed sensorless state estimation for induction motors: A moving horizon approach
Dong et al. Closed-loop voltage control of a parallel-plate MEMS electrostatic actuator
Pan et al. Advanced parameter identification for a linear-motor-driven motion system using disturbance observer
Li et al. An improved predictive current method for permanent magnet synchronous motors
Ranjbar et al. Adaptive sliding mode control of MEMS AC voltage reference source
Pandey et al. A newborn anti-windup scheme based on state prediction of fractional integrator for variable speed motor
Wang et al. Fractional-order Identification and Control of Permanent Magnet AC Servo System
Sueur Disturbance rejection with derivative state feedback

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17920247

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17920247

Country of ref document: EP

Kind code of ref document: A1