Subspace Identification for Closed-Loop Systems With Unknown Deterministic Disturbances

This paper presents a subspace identification method for closed-loop systems with unknown deterministic disturbances.To deal with the unknown deterministic disturbances,two strategies are implemented to construct the row space that can be used to approximately represent the unknown deterministic disturbances using the trigonometric functions or Bernstein polynomials depending on whether the disturbance frequencies are known.For closed-loop identification,CCF-N4SID is extended to the case with unknown deterministic disturbances using the oblique projection.In addition,a proper Bernstein polynomial order can be determined using the Akaike information criterion(AIC)or the Bayesian information criterion(BIC).Numerical simulation results demonstrate the effectiveness of the proposed identification method for both periodic and aperiodic deterministic disturbances.

PEMFC Fractional-order Subspace Identification Model

A proton exchange membrane fuel cell(PEMFC)is a new type of hydrogen fuel cell that plays an indispensable role in an energy network.However,the multivariable and fractional-order characteristics of PEMFC make it difficult to establish a practical model.Herein,a fractional-order subspace identification model based on the adaptive monarch butterfly optimization algorithm with opposition-based learning(ALMBO)algorithm is proposed for PEMFC.Introducing the fractional-order theory into the subspace identification method by adopting a Poisson filter for with input and output data,a weight matrix is proposed to improve the identification accuracy.Additionally,the ALMBO algorithm is employed to optimize the parameters of the Poisson filter and fractional order,which introduces an opposition-based learning strategy into the migration operator and incorporates adaptive weights to improve the optimization accuracy and prevent falling into a locally optimal solution.Finally,the PEMFC fractional-order subspace identification model is established,which can accurately describe the dynamic process of PEMFC.

An Extended Closed-loop Subspace Identification Method for Error-in-variables Systems

A closed-loop subspace identification method is proposed for industrial systems subject to noisy input-output observations, known as the error-in-variables (EIV) problem. Using the orthogonal projection approach to eliminate the noise influence, consistent estimation is guaranteed for the deterministic part of such a system. A strict proof is given for analyzing the rank condition for such orthogonal projection, in order to use the principal component analysis (PCA) based singular value decomposition (SVD) to derive the extended observability matrix and lower triangular Toeliptz matrix of the plant state-space model. In the result, the plant state matrices can be retrieved in a transparent manner from the above matrices. An illustrative example is shown to demonstrate the effectiveness and merits of the proposed subspace identification method.

IMPROVED COVARIANCE DRIVEN BLIND SUBSPACE IDENTIFICATION METHOD

An improved covariance driven subspace identification method is presented to identify the weakly excited modes. In this method, the traditional Hankel matrix is replaced by a reformed one to enhance the identifiability of weak characteristics. The robustness of eigenparameter estimation to noise contamination is reinforced by the improved Hankel matrix, in combination with component energy index （CEI） which indicates the vibration intensity of signal components, an alternative stabilization diagram is adopted to effectively separate spurious and physical modes. Simulation of a vibration system of multiple-degree-of-freedom and experiment of a frame structure subject to wind excitation are presented to demonstrate the improvement of the proposed blind method. The performance of this blind method is assessed in terms of its capability in extracting the weak modes as well as the accuracy of estimated parameters. The results have shown that the proposed blind method gives a better estimation of the weak modes from response signals of small signal to noise ratio （SNR）and gives a reliable separation of spurious and physical estimates.

Identification of Continuous-Time Linear Aircraft Models Using Subspace Identification

This paper considers the possibility of estimating continuous-time linear aircraft models by using the subspace identification The outline of the subspace identification for estimating the continuous-time linear model is first mentioned. The identification simulation in which the subspace identification is then applied to the modeling of aircraft in the lateral motion is performed. As a result, the continuous-time linear aircraft model is appropriately estimated by the subspace identification if the amount of the measurement noise is not so much.

Full-parameter constrained parsimonious subspace identification with steady-state information for DC–DC converters

A full-parameter constrained parsimonious subspace identification method that incorporates the steady-state a priori infor-mation of the system is proposed to model the DC-DC converters.A parsimonious model with fewer parameters is used to represent the system,and then an optimal weighted methods is used to estimate the system parameters matrices by taking into account both dynamical data and steady-state data.Compared with traditional data-driven methods for DC-DC convert-ers,the subspace-based method can simultaneously estimate model structure and parameter with appropriate computational complexity.Moreover,compared with the traditional full-parameter constrained subspace approach,the proposed algorithm can accurately estimate the system parameters with a smaller variance.The experimental results on a DC-DC synchronous buck converter verify the effectiveness and superiority of the proposed method.

N4SID and MOESP Algorithms to Highlight the Ill-conditioning into Subspace Identification

In this paper,an analysis for ill conditioning problem in subspace identifcation method is provided.The subspace identifcation technique presents a satisfactory robustness in the parameter estimation of process model which performs control.As a frst step,the main geometric and mathematical tools used in subspace identifcation are briefly presented.In the second step,the problem of analyzing ill-conditioning matrices in the subspace identifcation method is considered.To illustrate this situation,a simulation study of an example is introduced to show the ill-conditioning in subspace identifcation.Algorithms numerical subspace state space system identifcation(N4SID)and multivariable output error state space model identifcation(MOESP)are considered to study,the parameters estimation while using the induction motor model,in simulation(Matlab environment).Finally,we show the inadequacy of the oblique projection and validate the efectiveness of the orthogonal projection approach which is needed in ill-conditioning;a real application dealing with induction motor parameters estimation has been experimented.The obtained results proved that the algorithm based on orthogonal projection MOESP,overcomes the situation of ill-conditioning in the Hankel s block,and thereby improving the estimation of parameters.

A Hybrid Coordinated Design Method for Power System Stabilizer and FACTS Device Based on Synchrosqueezed Wavelet Transform and Stochastic Subspace Identification

The occurrence of low-frequency electromechanical oscillations is a major problem in the effective operation of power systems. The scrutiny of these oscillations provides substantial information about power system stability and security. In this paper, a new method is introduced based on a combination of synchrosqueezed wavelet transform and the stochastic subspace identification (SSI) algorithm to investigate the low-frequency electromechanical oscillations of large-scale power systems. Then, the estimated modes of the power system are used for the design of the power system stabilizer and the flexible alternating current transmission system (FACTS) device. In this optimization problem, the control parameters are set using a hybrid approach composed of the Prony and residual methods and the modified fruit fly optimization algorithm. The proposed mode estimation method and the controller design are simulated in MATLAB using two test case systems, namely IEEE 2-area 4-generator and New England-New York 68-bus 16-generator systems. The simulation results demonstrate the high performance of the proposed method in estimation of local and inter-area modes, and indicate the improvements in oscillation damping and power system stability.

Nonlinear System Identification Using Methods for Subspace

The identification of Wiener systems has been an active research topic for years. A Wiener system is a series connection of a linear dynamic system followed by a static nonlinearity. The difficulty in obtaining a representation of the Wiener model is the need to estimate the nonlinear function from the input and output data, without the intermediate signal availability. This paper presents a methodology for the nonlinear system identification of a Wiener type model, using methods for subspaces and polynomials of Chebyshev. The subspace methods used are MOESP （multivariable output-error state space） and N4SID （numerical algorithms for subspace state space system identification）. A simulated example is presented to compare the performance of these algorithms.

PEMFC Identification Based on a Fractional-Order Hammerstein State-Space Model with ADE-BH Optimization

Considering the fractional-order and nonlinear characteristics of proton exchange membrane fuel cells(PEMFC),a fractional-order subspace identification method based on the ADE-BH optimization algorithm is proposed to establish a fractional-order Hammerstein state-space model of PEMFCs.Herein,a Hammerstein model is constructed by connecting a linear module and a nonlinear module in series to precisely depict the nonlinear property of the PEMFC.During the modeling process,fractional-order theory is combined with subspace identification,and a Poisson filter is adopted to enable multi-order derivability of the data.A variable memory method is introduced to reduce computation time without losing precision.Additionally,to improve the optimization accuracy and avoid obtaining locally optimum solutions,a novel ADEBH algorithm is employed to optimize the unknown parameters in the identification method.In this algorithm,the Euclidean distance serves as the theoretical basis for updating the target vector in the absorption-generation operation of the black hole(BH)algorithm.Finally,simulations demonstrate that the proposed model has small output error and high accuracy,indicating that the model can accurately describe the electrical characteristics of the PEMFC process.

Novel Model of Proton Exchange Membrane Fuel Cell with Predictive Control Using Hildreth Algorithm Based on Fractional-order Dynamic Model

Considering the multivariable and fractional-order characteristics of proton exchange membrane fuel cells(PEMFCs),a fractional-order subspace identification method(FOSIM)is proposed in this paper to establish a fractionalorder state space(FOSS)model,which can be expressed as a multivariable configuration with two inputs,hydrogenflow rate and stack current,and two outputs,cell voltage and power.Based on this model,a novel constrained optimal control law named the Hildreth model predictive control(H-MPC)strategy is created,which employs a Hildreth quadratic programming algorithm to adjust the output power of fuel cells through adaptively regulating hydrogen flow and stack current.dSPACE semi-physical simulation results demonstrate that,compared with proportional-integral-derivative and quadratic programming MPC(QP-MPC),the proposed H-MPC exhibits better tracking ability and strong robustness against variations of PEMFC power.

Automatic modal parameter identification of high arch dams:feasibility verification

Modal parameters, including fundamental frequencies, damping ratios, and mode shapes, could be used to evaluate the health condition of structures. Automatic modal parameter identification, which plays an essential role in realtime structural health monitoring, has become a popular topic in recent years. In this study, an automatic modal parameter identification procedure for high arch dams is proposed. The proposed procedure is implemented by combining the densitybased spatial clustering of applications with noise(DBSCAN) algorithm and the stochastic subspace identification(SSI). The 210-m-high Dagangshan Dam is investigated as an example to verify the feasibility of the procedure. The results show that the DBSCAN algorithm is robust enough to interpret the stabilization diagram from SSI and may avoid outline modes. This leads to the proposed procedure obtaining a better performance than the partitioned clustering and hierarchical clustering algorithms. In addition, the errors of the identified frequencies of the arch dam are within 4%, and the identified mode shapes are in agreement with those obtained from the finite element model, which implies that the proposed procedure is accurate enough to use in modal parameter identification. The procedure is feasible for online modal parameter identification and modal tracking of arch dams.

Investigation of Dynamic Multivariate Chemical Process Monitoring

Chemical process variables are always driven by random noise and disturbances. The closed-loop con-trol yields process measurements that are auto and cross correlated. The influence of auto and cross correlations on statistical process control (SPC) is investigated in detail by Monte Carlo experiments. It is revealed that in the sense of average performance, the false alarms rates (FAR) of principal component analysis (PCA), dynamic PCA are not affected by the time-series structures of process variables. Nevertheless, non-independent identical distribution will cause the actual FAR to deviate from its theoretic value apparently and result in unexpected consecutive false alarms for normal operating process. Dynamic PCA and ARMA-PCA are demonstrated to be inefficient to remove the influences of auto and cross correlations. Subspace identification-based PCA (SI-PCA) is proposed to improve the monitoring of dynamic processes. Through state space modeling, SI-PCA can remove the auto and cross corre-lations efficiently and avoid consecutive false alarms. Synthetic Monte Carlo experiments and the application in Tennessee Eastman challenge process illustrate the advantages of the proposed approach.

Orthogonal projection based subspace identification against colored noise

In this paper, a bias-eliminated subspace identification method is proposed for industrial applications subject to colored noise. Based on double orthogonal projections, an identification algorithm is developed to eliminate the influence of colored noise for consistent estimation of the extended observability matrix of the plant state-space model. A shift-invariant approach is then given to retrieve the system matrices from the estimated extended observability matrix. The persistent excitation condition for consistent estimation of the extended observability matrix is analyzed. Moreover, a numerical algorithm is given to compute the estimation error of the estimated extended observability matrix. Two illustrative examples are given to demonstrate the effectiveness and merit of the proposed method.

Model variables identification of a gas turbine using a subspace approach based on input/output data measurements

This paper deals with a state model identification of a gas turbine used for gas transport, using a subspace approach of the state space model. This method provides a reliable and robust state representation of the model, taking advantage of its benefits in the control, monitoring, and supervision of this machine. The model for each variable is set so that the state matrices associated with the gas turbine model are determined from their real input/output data. The comparison of the obtained identification results with those of the actual turbine operation serves to validate the proposed model in this work. This numerical algorithm of the subspace identification method is full of information and more accurate in terms of residual modeling error, and expresses a very high level of confidence in the identified turbine system dynamics. Hence, the controllability and observability tests of turbine operation for different input/output variables allowed to validate the real-time operating stability of the turbine.

VIBRATION CONTROL OF FLUID-FILLED PRISMATIC SHELL WITH ACTIVE CONSTRAINED LAYER DAMPING TREATMENTS

Active constrained layer damping （ACLD） combines the simplicity and reliability of passive damping with the light weight and high efficiency of active actuators to obtain high damping over a wide frequency band. A fluid-filled prismatic shell is set up to investigate the validity and efficiency of ACLD treatments in the case of fluid-structure interaction. By using state subspace identification method, modal parameters of the ACLD system are identified and a state space model is established subsequently for the design of active control laws. Experiments are conducted to the fluid-filled prismatic shell subjected to random and impulse excitation, respectively, For comparison, the shell model without fluid interaction is experimented as well. Experimental results have shown that the ACLD treatments can suppress vibration of the fluid-free and fluid-filled prismatic shell effectively. Under the same control gain, vibration attenuation is almost the same in both cases.

Monitoring the Modal Frequency and Velocity Variations of Bridges under Ambient Excitation

We conducted a long-term monitoring experiment on the Lutuanxilu Bridge located in Changping District of Beijing, employing our recently developed real-time bridge monitoring system based on the Guralp CMG-6TD broadband seismometer. We identified the modal parameters with the stochastic subspace identification( SSI) algorithm,and continuously monitored the temporal velocity variation with coda wave interferometry.The results show that:( 1) the highly sensitive Guralp CMG-6TD broadband seismometer,which records the three-component vibration signal within broad frequency range,is well suited for long-term bridge health monitoring.( 2) With the continuous vibration signal from ambient excitation,the stochastic subspace algorithm can robustly identify the low-order modal parameters and the coda wave interferometry can accurately monitor the tiny velocity variation.( 3) The elastic modulus of bridge materials changes significantly associated with varying temperature,leading to diurnal velocity variation with amplitude of approximately 1%. The velocity variation shows strong negative correlation with temperature fluctuation. Meanwhile,the modal frequencies remain quite stable,suggesting that the velocity variation may be a more sensitive quantitative damage index.( 4) While the modal frequencies reflect the integrated health status of the bridge,the velocity variation can be utilized to monitor the local elastic modulus. Therefore,it is crucial for bridge health monitoring to continuously monitor the two key damage indexes under ambient excitation.

Research on the Prediction of Load Regulation Capacity for Supercritical Thermal Power Unit

Both the modeling and the load regulation capacity prediction of a supercritical power plant are investigated in this paper. Firstly, an indirect identification method based on subspace identification method is proposed. The obtained identification model is verified by the actual operation data and the dynamic characteristics of the system are well reproduced. Secondly, the model is used to predict the load regulation capacity of thermal power unit. The power, main steam pressure, main steam temperature and other parameters are simulated respectively when the unit load is going up and down. Under the actual constraints, the load regulation capacity of thermal power unit can be predicted quickly.

Subspace-based identification of discrete time-delay system

We investigate the identification problems of a class of linear stochastic time-delay systems with unknown delayed states in this study. A time-delay system is expressed as a delay differential equation with a single delay in the state vector. We first derive an equivalent linear time-invariant(LTI) system for the time-delay system using a state augmentation technique. Then a conventional subspace identification method is used to estimate augmented system matrices and Kalman state sequences up to a similarity transformation. To obtain a state-space model for the time-delay system, an alternate convex search(ACS) algorithm is presented to find a similarity transformation that takes the identified augmented system back to a form so that the time-delay system can be recovered. Finally, we reconstruct the Kalman state sequences based on the similarity transformation. The time-delay system matrices under the same state-space basis can be recovered from the Kalman state sequences and input-output data by solving two least squares problems. Numerical examples are to show the effectiveness of the proposed method.

Identification of reduced-order model for an aeroelastic system from flutter test data

Recently, flutter active control using linear parameter varying（LPV） framework has attracted a lot of attention. LPV control synthesis usually generates controllers that are at least of the same order as the aeroelastic models. Therefore, the reduced-order model is required by synthesis for avoidance of large computation cost and high-order controller. This paper proposes a new procedure for generation of accurate reduced-order linear time-invariant（LTI） models by using system identification from flutter testing data. The proposed approach is in two steps. The well-known poly-reference least squares complex frequency（p-LSCF） algorithm is firstly employed for modal parameter identification from frequency response measurement. After parameter identification,the dominant physical modes are determined by clear stabilization diagrams and clustering technique. In the second step, with prior knowledge of physical poles, the improved frequencydomain maximum likelihood（ML） estimator is presented for building accurate reduced-order model. Before ML estimation, an improved subspace identification considering the poles constraint is also proposed for initializing the iterative procedure. Finally, the performance of the proposed procedure is validated by real flight flutter test data.