连续时间模型的闭环子空间辨识

本文对直接使用采样数据进行连续系统的闭环子空间辨识问题进行了研究．将线性滤波方法与基于主 元分析的子空间辨识相结合,利用参考输入或者外部激励信号的高阶滤波变换的正交投影变量作为辅助变量,提出 了一种新的连续时间系统闭环子空间辨识算法．数值仿真表明了与其他算法相比,本文提出的算法具有很好的辨识 效果．     

Improved closed-loop subspace identification with prior information

It has been proven that combining open-loop subspace identification with prior information can promote the accuracy of obtaining state-space models. In this study, prior information is exploited to improve the accuracy of closed-loop subspace identification. The proposed approach initially removes the correlation between future input and past innovation, a significant obstacle in closed-loop subspace identification method. Then, each row of the extended subspace matrix equation is considered an optimal multi-step ahead predictor and prior information is expressed in the form of equality constraints. The constrained least squares method is used to obtain improved results, so that the accuracy of the closed-loop subspace can be enhanced. Simulation examples are provided to demonstrate the effectiveness of the proposed algorithm.     

Subspace identification using instrumental variable techniques

Subspace-based algorithms for system identification have lately been suggested as alternatives to more traditional techniques. Variants of the MOESP type of subspace algorithms are in addition to open-loop identification applicable to closed-loop and errors-in-variables identification. In this paper, a new instrumental variable approach to subspace identification is presented. It is shown how existing MOESP-algorithms can be derived within the proposed framework, simply by changing instruments and weighting matrices. A noteworthy outcome of the analysis is that an improvement of an existing MOESP method for errors-in-variables identification can be proposed.     

基于正交分解的递推子空间闭环辨识方法

为实现闭环系统在线辨识,提出递推正交分解闭环子空间辨识方法(RORT)。首先,根据闭环系统状态空间模型和数据间投影关系,构建确定-随机模型,并利用GIVENS变换实现投影向量的递推QR分解;然后,引入带遗忘因子的辨识算法,构建广义能观测矩阵的递推更新形式,以减少子空间辨识算法中QR分解和SVD分解的计算量;最后,针对某型号陀螺仪闭环系统进行实验。实验结果表明, RORT法的辨识拟合度高于91%,能够对陀螺仪闭环系统模型参数进行在线监测。     

基于输入扩张的闭环系统子空间辨识及其强一致性分析

针对闭环条件下的子空间辨识问题, 结合线性代数和几何学的基本概念, 将输入输出误差序列包含至输入子空间中, 基于输入扩张的状态空间构造方法, 提出一种新的闭环辨识算法；解决开环算法应用于闭环系统辨识时产生有偏估计, 甚至不能正确辨识的问题；实现闭环条件下对系统状态空间矩阵的强一致估计, 并理论证明该辨识算法的强一致性；最后通过仿真实例验证本算法的有效性.     

基于辅助变量的闭环系统子空间辨识

提出一种基于辅助变量的子空间辨识方法,适用于控制器信息未知以及参考输入已知的闭环系统参数辨识.通过将输入-输出数据块正交投影到辅助变量的行空间,直接得到扩展观测矩阵垂空间的估计.由此可从闭环系统中提取出对象模型信息,同时由SVD分解得到扩展观测矩阵与下三角Toeplitz矩阵的估计.给出了系统参数矩阵、噪声矩阵的计算方法.将所提出的子空间辨识方法应用于闭环动态的系统参数估计,其结果表明了该方法的有效性.     

基于两步方法的闭环子空间辨识算法

闭环辨识算法具有广泛的工程应用前景,而子空间方法近年来应用于多个领域中,但子空间方法无法直接应用于闭环辨识,因此研究闭环子空间辨识算法具有重要意义.两步方法可用于辨识闭环系统,但计算量巨大,推导复杂,需要进一步改进.针对这种情况,提出了一种改进算法,使用将两步方法与正交投影相结合的方法,并利用QR分解实现,直接构建虚拟信号序列,大大减少了计算量,最后使用PI-MOESP辨识算法辨识模型.仿真实验将该算法与其他子空间辨识算法相比较,显示出该算法的有效性及计算量的显著减少.     

一种无外部持续激励的子空间闭环辨识算法

传统的闭环系统可辨识性受到外部激励和控制器结构的限制,然而在实际辨识应用中,由于过程操作条件的限制或者经济原因不希望存在外部持续激励。讨论了在无外部激励下基于快采样的多变量子空间模型闭环可辨识性,提出了无外部持续激励条件下的闭环快采样子空间模型辨识方法,在常规采样的可辨识性条件不满足或部分满足的情况下,通过对闭环系统控制对象的输入输出变量快采样来实现对象模型辨识,并通过仿真实例验证了基于快采样的辨识算法的有效性。     

Closed-loop subspace-based identification algorithm using third-order cumulants

The problem of closed-loop system identification for coloured noise system without any knowledge of feedback controller is considered. We develop a solution to this problem in the framework of subspace identification based on high-order cumulants. The key of the developed algorithm is using the properties that the third-order cumulants are insensitive to any coloured Gaussian noises. By post-multiplying a suitable instrumental variable to the noise terms, the cross third-order cumulants are constructed that become zero when the noises are Gaussian distributed, and meanwhile the column rank of extended observability matrix is maintained. Thus, the standard subspace identification algorithms can be extended to closed-loop system corrupted by arbitrary coloured noises. A numerical simulation is presented to demonstrate the proposed algorithm.     

递推子空间模型辨识算法的研究与仿真

为了很好的解决在线辨识系统模型问题,在对子空间模型辨识研究的基础上,结合递推最小二乘算法和子空问状态辨识方法。推导了子空间状态辨识的递推算法。该算法不仅解决了在线辨识问题,而且算法简单,计算方便,很好地克服了在线辨识时子空间矩阵维数的变化问题。经仿真研究表明,该递推算法克服了一次完成算法在大批量数据运算时,耗时大,专用内存多的缺点,而且对于测量和过程均有噪声干扰的多输入多输出系统,有很好的辨识效果,有较为广阔的应用前景。     

一种子空间辨识的广义预测控制算法

针对广义预测控制(GPC)模型中输入输出数据可能存在噪声和系统先验结构信息未知导致的难于辨识问题,提出了一种子空间辨识的广义预测控制算法。该算法采用变遗忘因子的子空间辨识方法,按照预测优化值与参考输出值的误差构造变遗忘因子,调整采集数据权重,进行在线辨识以提高灵敏度和控制效果。实验结果验证了所提出算法的有效性。     

数据驱动闭环子空间预测控制方法研究与应用

提出一种完全数据驱动的闭环子空间辨识及预测控制器设计方法. 该方法完全由闭环系统的输入输出数据辨识子空间矩阵, 通过子空间矩阵的拆分, 排除了与扰动相关的模型输入, 进而获取子空间矩阵参数的无偏估计; 将辨识得到的闭环系统子空间矩阵描述直接作为预测模型, 设计预测控制器; 将其应用于某钢铁集团焦炉炭化室压力控制系统, 取得了良好的控制效果.

     

Realizations of BELS as WIV method in both direct and indirect closed-loop system identification

The bias eliminated least squares (BELS) method, which is known as efficient for unknown parameter estimation of transfer function in the correlated noise case, has been developed and applied effectively to the closed-loop system identification. In this paper, under the general settings, the realizations of the BELS method as a weighted instrumental variables (WIV) method in both direct and indirect closed-loop system identification are established through constructing an appropriate weighting matrix in the WIV method. The constructed structures are similar in both cases, which reveals that all the proof procedures of the two realizations are the same. Thus, the unified realizations of the BELS as the WIV method for the closed-loop system identification can be built. A simulation example is given to validate our theoretical analysis. Supported by the National Natural Science Foundation of China for Distinguished Young Scholars (Grant No. 60625104), the Ministerial Foundation of China (Grant No. A2220060039), and the Fundamental Research Foundation of BIT (Grant No. 1010050320810)     

基于新息估计和正交投影的闭环子空间模型辨识

针对闭环控制系统提出一种基于新息估计和正交投影的闭环子空间模型辨识方法.首先采用最小二乘法对VARX模型（Vector autoregressive with exogenous inputs model）进行计算得到新息估计值,然后通过将由观测输入输出数据构造的Hankel矩阵正交投影到新息数据的正交补空间以消除噪声影响,从而在无噪声的输入输出数据奇偶空间中提取得到扩展可观测矩阵和下三角形Toeplitz矩阵.最后采用平移变换法得到系统矩阵.对该算法严格分析和证明了实现一致估计的条件.通过仿真实例验证了本文方法的有效性和优越性.     

基于子空间方法的精馏塔系统辨识建模研究

子空间辨识方法作为一种有效的针对多输入-多输出系统(MIMO)的辨识建模方法近年来受到广泛的重视.目前主要采用的子空间辨识算法只能适用于白噪声环境,而实际的工业现场数据很多是受到较大有色噪声干扰的.针对问题采用了一种新的子空间辨识算法,利用马尔可夫参数用于处理随机性部分,同时引入辅助变量用以去除噪声的干扰,能够适用于存在较大有色噪声干扰情况下的辨识建模,并可得到对象的无偏模型,建模的精度优于通常所采用的子空间辨识算法.通过对精馏塔仿真模型的辨识结果证明了该方法的可行性和有效性,以及在实际工业过程对象建模中良好的应用前景.     

分数阶系统时域子空间辨识

研究了分数阶系统的时域辨识问题,给出了一种新的分数阶系统时域子空间辨识算法.当分数阶微分阶次已知时,通过计算输入输出信号的分数阶微分,构造新的输入输出数据方程对系统的参数进行子空间辨识.当分数阶微分阶次未知时,通过代价函数将阶次辨识问题转化为参数寻优问题.采用Poisson滤波器有效避免了在计算分数阶微分时输入输出信号必须高阶可导的问题.通过分析给出了权矩阵的选取方式,提高了时域子空间辨识结果的精度.数值仿真结果表明了该算法的有效性.     

Subspace identification with non-steady Kalman filter parameterization

Most existing subspace identification methods use steady-state Kalman filter (SKF) in parameterization, hence, infinite data horizons are implicitly assumed to allow the Kalman gain to reach steady state. However, using infinite horizons requires collecting infinite data which is unrealistic in practice. In this paper, a subspace framework with non-steady state Kalman filter (NKF) parameterization is established to provide exact parameterization for finite data horizon identification problems. Based on this we propose a novel subspace identification method with NKF parameterization which can handle closed-loop data and avoid assumption on infinite horizons. It is shown that with finite data, the proposed parameterization method provides more accurate and consistent solutions than existing SKF based methods. The paper also reveals why it is often beneficial in practice to estimate a bank of ARX models over a single ARX model.     

Closed-loop data-driven simulation

Closed-loop data-driven simulation refers to the problem of finding the set of all responses of a closed-loop system to a given reference signal directly from an input/output trajectory of the plant and a representation of the controller. Conditions under which the problem has a solution are given and an algorithm for computing the solution is presented. The problem formulation and its solution are in the spirit of the deterministic subspace identification algorithms, i.e. in the theoretical analysis of the method, the data is assumed exact (noise free). The results have applications in data-driven control, e.g. testing controller's performance directly from closed-loop data of the plant in feedback with possibly different controller.     

Consistency analysis of some closed-loop subspace identification methods

We study statistical consistency of two recently proposed subspace identification algorithms for closed-loop systems. These algorithms may be seen as implementations of an abstract state-space construction procedure described by the authors in previous work on stochastic realization of closed-loop systems. A detailed error analysis is undertaken which shows that both algorithms are biased due to an unavoidable mishandling of initial conditions which occurs in closed-loop identification. Instability of the open loop system may also be a cause of trouble.     

分数阶时滞系统的频域子空间辨识

研究了利用频率响应数据辨识分数阶时滞系统子空间模型的问题,给出了一种差分进化算法与频域子空间方法相结合的辨识算法.利用差分进化算法搜索最优分数微分阶次和时滞参数,而对于固定的分数微分阶次和时滞,则采用分数阶频域子空间辨识方法得到状态空间模型.通过仿真算例验证了该算法的有效性.