A. Poulimenos, M. Spiridonakos, and S. Fassois

Size: px
Start display at page:

Download "A. Poulimenos, M. Spiridonakos, and S. Fassois"

Transcription

1 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS: AN OVERVIEW AND COMPARISON A. Poulimenos, M. Spiridonakos, and S. Fassois DEPARTMENT OF MECHANICAL & AERONAUTICAL ENGINEERING, GR 265 PATRAS, GREECE sms ISMA26 International Conference on Noise and Vibration Engineering Leuven, Belgium, September 26

2 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 1 TALK OUTLINE 1. INTRODUCTION 2. THE PRESENT PARADIGM (CASE STUDY) 3. THE GENERAL TARMA MODEL REPRESENTATION 4. MODEL PARAMETER ESTIMATION 5. MODEL STRUCTURE ESTIMATION 6. NON-STATIONARY VIBRATION MODELLING RESULTS 7. CONCLUDING REMARKS

3 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 2 1. INTRODUCTION The General Problem Modelling and analysis of non-stationary vibration. Problem Characteristics Non-stationary stochastic signals are characterized by time-varying statistics require time-frequency methods for their analysis. Mechanical Vibration Acceleration (cm/sec 2 ) Output signal (location 3) Time (sec)

4 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 3 Problem Significance Non-stationary random vibration is commonly encountered in practice: Seismic & structural systems Sea vessels Rotating machinery Automotive & aircraft systems Robotic devices and so on Methods for Non-stationary Random Vibration Modelling and Analysis Non-parametric (Hammond & White 1996; Cohen 1994; Spanos & Failla 25) Parametric (Poulimenos & Fassois 26; Niedzwiecki 2; Kitagawa & Gersch 1996; Owen et al. 21; Cooper & Worden 2) Spectrogram (STFT) Cohen class of distributions Wavelet-based methods TARMA & State-space Unstructured parameter evolution Stochastic parameter evolution Deterministic parameter evolution

5 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 4 Why Parametric Methods? They offer a number of advantages over non-parametric counterparts: Representation parsimony Improved accuracy and resolution Improved tracking of the time-varying dynamics Flexibility in analysis Aims of the Work 1. Critical overview of parametric time-domain methods for non-stationary random vibration modelling and analysis. 2. Comparative assessment of the methods via Monte-Carlo experiments (simulated non-stationary vibration signal with precisely known characteristics).

6 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 5 2. THE PRESENT PARADIGM (CASE STUDY) THE PROBLEM: Modeling, analysis and prediction of non-stationary vibration. The Underlying System m 3 k 3 (t) c 3 x 3 Time-varying stiffnesses k 2 (t) and k 3 (t): k i (t) = k i, + k i,1 sin(2πt/p i,1 ) + k i,2 sin(2π/p i,2 ) m 2 k 2 (t) m 1 k 1 c 2 x 2 x 1 r(t) Symbol Value m 1.5 kg m kg m 3 1. kg c 2.5 N/(m/s) c 3.3 N/(m/s) k 1 3 N/m k 2 (t) k 2, = 1 (N/m), k 2,1 = 6, k 2,2 = 2 (N/m) P 2,1 = 17, P 2,2 = 85 (s) k 3 (t) k 3, = 12 (N/m), k 3,1 = 72, k 3,2 = 24 (N/m) P 3,1 = , P 3,2 = (s)

7 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 6 System Properties 6 5 ω n3 Frequency (Hz) ω n2 1 ω n1 The Non-stationary Vibration Signal Time (sec) STFT Displacement Time (sec)

8 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 7 3. THE GENERAL TARMA MODEL REPRESENTATION A TARMA(n a, n c ) model is of the form: x[t] + n a a i [t] x[t i] i=1 }{{} AR part = e[t] + n c c i [t] e[t i], i=1 }{{} MA part t t, e[t] NID(, σ 2 e[t]) t : normalized discrete time n a, n c : orders of the AR/MA polynomials x[t] : the non-stationary response signal a i [t], c i [t] : AR/MA time-varying parameters e[t] : innovations (residual) sequence Classes of Parametric Methods Parameter evolution Representation parsimony Dynamics evolution UNSTRUCTURED low slow STOCHASTIC low slow/medium DETERMINISTIC high slow/medium/fast

9 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 8 4. MODEL PARAMETER ESTIMATION Estimation of the AR, MA parameter vector and the residual variance at all time instants for selected model form and structure [ ] T, θ[t] = a 1 [t]... a na [t]. c }{{} 1 [t]... c nc [t] σ 2 }{{} e [t], (t = 1,..., N) AR parameters MA parameters }{{} residual variance Unstructured Parameter Evolution (UPE-TARMA) Models The AR/MA parameters do not have a structured time-dependency, and are free to change with time. Their estimation may be achieved either by locally stationary or recursive methods. Locally stationary methods [Short Time ARMA (ST-ARMA) estimation] Conventional, stationary ARMA modelling to successive, short (approximately stationary), time segments. Acceleration a) b) Time (sec)

10 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 9 Recursive methods The AR/MA parameter vector estimate θ[t] at each time instant t is recursively updated at the time instant t + 1 that the next signal sample becomes available. 1 Acceleration Time (sec) t 1 t ˆθ[t] = ˆθ[t 1] + k[t] ê[t t 1] }{{} prediction error θ[t] = [ a 1 [t]...a na [t]. c 1 [t]... c nc [t] ] T Exponentially Weighted Prediction Error criterion: ˆ θ[t] = arg min θ[t] t λ t τ e 2 [τ, θ τ 1 ] τ=1 The RML-TARMA method is presently used (Ljung 1999).

11 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 1 RML-TARMA Estimation Estimator update: ˆθ[t] = ˆθ[t 1] + k[t]ê[t t 1] Prediction error: ê[t t 1] = x[t] ˆx[t t 1] = x[t] φ T [t]ˆθ[t 1] Gain: P [t 1]ψ[t] k[t] = λ + ψ T [t]p [t 1]ψ[t] Covariance update: P [t] = 1 ( P [t 1] P [t ) 1]ψ[t]ψT [t]p [t 1] λ λ + ψ T [t]p [t 1]ψ[t] Filtering: ψ[t] + ĉ 1 [t 1]ψ[t 1] ĉ nc [t 1]ψ[t n c ] = φ[t] A-posteriori error: ê[t t] = x[t] φ T [t]ˆθ[t] φ[t] = [ ] T x[t 1]... x[t n α ]. ê[t 1 t 1]... ê[t n c t n c ] Initialization: ˆθ[] =, P [] = αi with α designating a large positive number. The signal and a-posteriori error initial values are set to zero.

12 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 11 Stochastic Parameter Evolution (SP-TARMA) Models (Kitagawa & Gersch 1996) The AR/MA parameters are considered as being random variables allowed to change with time, with their evolution subject to stochastic smoothness constraints: κ α i [t] = w αi [t] κ c i [t] = w ci [t], ( = 1 B).2.3 α i Time (sec) SP-TARMA models may be set into linear state space form: t 2, t 1, t α i [t] = 2 α i [t 1] α i [t 2] + w αi [t] (κ = 2) z[t] = F z[t 1] + G w[t], x[t] = h T [t,z t 1 ] z[t] + e[t] z[t] = [ a 1 [t]... a na [t] c 1 [t]...c nc [t].....a 1 [t κ+1]...a na [t κ+1] c 1 [t κ+1]... c nc [t κ+1] ] T Estimation of the state vector z[t] at each time instant may be achieved by the Kalman Filter. A smoothed estimate ẑ[t N] of the state vector may be obtained via a backward smoothing algorithm.

13 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 12 Kalman Filter and Backward Smoothing for the Estimation of SP-TARMA Models (normalized form) Time update (prediction): State prediction: ẑ[t t 1] = Fẑ[t 1 t 1] Prediction error: ê[t t 1] = x[t] h T [t]ẑ[t t 1] Covariance update: P [t t 1] = F P [t 1 t 1]F T + G Q[t]G T Observation update (filtering): Gain: k[t] = P [t t 1]h[t] State update: ẑ[t t] = ẑ[t t 1] + k[t]ê[t t 1] Covariance update: P [t t] = ( I k[t]h T [t] ) P [t t 1] ( ) 1 h T [t] P [t t 1]h[t] + 1 Smoothing: A[t] = P [t t]f T P 1 [t + 1 t] ẑ[t N] = ẑ[t t] + A[t] (ẑ[t + 1 N] ẑ[t + 1 t]) ( ) P [t N] = P [t t] + A[t] P [t + 1 N] P [t + 1 t] A T [t] P [t t] = P [t t] σe[t], P [t t 1] 2 = P [t t 1], Q[t] σe[t] 2 = Q[t] σe[t] = σ2 w[t] I 2 σ }{{} e[t] 2 na ν[t] Initialization: ˆθ[] =, P [] = αi with α designating a large positive number.

14 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 13 Deterministic Parameter Evolution (FS-TARMA) Models (Grenier 1989, Poulimenos & Fassois 26) The AR/MA parameters and innovations variance are deterministic functions of time, belonging to specific functional subspaces: F AR = {Gba (1)[t],..., G ba (p a )[t]}, F MA = {Gbc (1)[t],..., G bc (p c )[t]}, F σ 2 e = {Gbs (1)[t],..., G bs (p s )[t]} α i Time (sec) G ba (1)[t] G ba (2)[t] G ba (2)[t] Time (sec) 2 2 G ba (4)[t] G ba (5)[t] G ba (6)[t] Time (sec) 1 Time (sec) a i [t] = 6 j=1 a i,j G ba (j)[t], a i,j = [ ] T Estimation of the model parameters a i,j, c i,j, s j (coefficients of projection), may be based upon a Prediction Error criterion (non-quadratic problem). The original PE problem may be tackled by: linear multistage methods: 2SLS method, P-A method recursive methods: Recursive Extended Least Squares (RELS) method

15 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 14 Multistage Identification of FS-TARMA Models The Two-Stage Least Squares (2SLS) method The Polynomial-Algebraic (P-A) method NO x[t] (t = 1,...,N) Inverse Function Estimation AR/MA Projection Coefficient Estimation î ˆϑ RSSn RSS n 1 ε RSS n 1 YES Innovations Variance Estimation ˆϑ, ŝ A truncated FS-TAR model is estimated through OLS. The FS-TARMA model is approximated by replacing the past, but not the current, values of the prediction error e[t, ϑ] by the obtained residual series e[t, î]. The estimation of the AR/MA coefficient of projection vector ϑ is then reduced to a quadratic optimization problem. NO x[t] (t = 1,...,N) Inverse Function Estimation Initial AR/MA Projection Coefficient Estimation Signal Filtering Final AR/MA Projection Coefficient Estimation RSSn RSS n 1 ε RSS n 1 YES Innovations Variance Estimation A truncated FS-TAR representation is estimated through OLS. Initial estimates of the AR and MA coefficients of projection are obtained based on the definition of the inverse function. A filtered signal x[t] is obtained by consecutive filtering operations. The final AR coefficients of projection are estimated based on the filtered signal. The final MA coefficient of projection estimates are obtained through deconvolution. ˆϑ, ŝ

16 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS MODEL STRUCTURE ESTIMATION Estimation of the model orders and additional structural parameters for a selected model class M = {n a, n c } M SP = {n a, n c, κ} M FS = {na, n c, F AR, F MA, F σ 2 e } n a, n c : AR/MA orders κ : smoothness constraints order F : functional subspace General approach: minimization of the Bayessian Information Criterion (BIC) or the log-likelihood function. BIC = [ N 2 ln 2π N ( ln(σe[t]) 2 + e2 [t] σe[t] 2 ) ] t=1 }{{} ln L(x N ) + lnn 2 d: the number of independently estimated model parameters d

17 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 16 Integer Optimization Scheme (Poulimenos and Fassois 23) Minimization of the BIC is achieved via a direct, hybrid, two-phase optimization scheme: PHASE I. Coarse ( global ) optimization. is applied to the complete search space aiming at locating promising regions. Approach: Genetic Algorithms PHASE II. Fine ( local ) optimization. operates in the neighborhood of each Phase I result aiming at locating the exact optimum point. Approach: Backward Regression

18 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS NON-STATIONARY VIBRATION MODELLING RESULTS Model structure selection: (a) TARMA(n,n) models (n = 2, 4,...) are estimated (b) MA order reduction is considered Criteria: BIC and log-likelihood function BIC BIC AR order MA order

19 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 18 Identified Models Model Class Identification Method Method Characteristics Identified Model Unstructured ST-ARMA M = 31; m = 1; PE estimation ST-ARMA(6,3) Parameter (Levenberg-Marquardt: Evolution termination rule ˆϑ i ˆϑ i 1 < 1 6 ) RML-TARMA λ =.978; initl. covariance α = 1 4 RML-TARMA(6,3) Stochastic SP-TARMA ELS-like estimation; α = 1 4 SP-TARMA(6,3) κ = 2 Parameter ˆν = Evolution SP-TARMA (smoothed) Deterministic FS-TARMA (2SLS) n i = 2; QR implementation of OLS FS-TARMA(6,3) [41,2,19] Parameter FS-TARMA (2SLS-PE) Evolution (Levenberg-Marquardt) FS-TARMA (RELS) initl. covariance α = 1 4 SP-TARMA(6, 3) models: α i [t] = 2 α i [t 1] α i [t 2] + w αi [t], c i [t] = 2 c i [t 1] c i [t 2] + w ci [t] FS-TARMA(6, 3) models: trigonometric functional subspaces

20 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 19 Model-Based Natural Frequency Estimates 6 : Theoretical : FS TARMA (2SLS PE) : FS TARMA (RELS) : SP TARMA (smoothed) ω n3 5 Frequency (Hz) ω n2 1 ω n Time (sec)

21 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 2 Model-Based Natural Frequency Estimates 6 : Theoretical : RML TARMA : SP TARMA ω n3 5 Frequency (Hz) ω n2 1 ω n Time (sec)

22 PARAMETRIC T IME -D OMAIN M ETHODS FOR N ON -S TATIONARY R ANDOM V IBRATION I DENTIFICATION AND A NALYSIS 21 Model-Based Frozen Vibration Analysis ST-ARMA(6,3) S TOCHASTIC M ECHANICAL S YSTEMS & AUTOMATION (SMSA) L ABORATORY RML-TARMA(6,3) U NIVERSITY OF PATRAS

23 PARAMETRIC T IME -D OMAIN M ETHODS FOR N ON -S TATIONARY R ANDOM V IBRATION I DENTIFICATION AND A NALYSIS 22 Model-Based Frozen Vibration Analysis SP-TARMA(6,3) S TOCHASTIC M ECHANICAL S YSTEMS & AUTOMATION (SMSA) L ABORATORY SP-TARMA(6,3) (smoothed) U NIVERSITY OF PATRAS

24 PARAMETRIC T IME -D OMAIN M ETHODS FOR N ON -S TATIONARY R ANDOM V IBRATION I DENTIFICATION AND A NALYSIS 23 Model-Based Frozen Vibration Analysis FS-TARMA(6,3) (2SLS) S TOCHASTIC M ECHANICAL S YSTEMS & AUTOMATION (SMSA) L ABORATORY FS-TARMA(6,3) (RELS) U NIVERSITY OF PATRAS

25 PARAMETRIC T IME -D OMAIN M ETHODS FOR N ON -S TATIONARY R ANDOM V IBRATION I DENTIFICATION AND A NALYSIS 24 Model-Based Frozen Vibration Analysis FS-TARMA(6,3) (2SLS-PE) S TOCHASTIC M ECHANICAL S YSTEMS & AUTOMATION (SMSA) L ABORATORY Theoretical U NIVERSITY OF PATRAS

26 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS 25 Comparative Results RSS/SSS (%) CPU time (normalized %) 1 2 1% %.15%.5%.86% 42.46% 17.57% ST ARMA RML RARMA SP TARMA SP TARMA (smoothed) FS TARMA (2SLS) FS TARMA (2SLS PE) FS TARMA (RELS) Parametric Methods Main Characteristics TARMA Representation Modelling/Prediction Vibration Computational Ease Method Parsimony Accuracy Analysis Simplicity of Use ST-ARMA n/a RML-TARMA SP-TARMA SP-TARMA (smoothed) FS-TARMA (2SLS) FS-TARMA (2SLS-PE) FS-TARMA (RELS)

27 PARAMETRIC TIME-DOMAIN METHODS FOR NON-STATIONARY RANDOM VIBRATION IDENTIFICATION AND ANALYSIS CONCLUDING REMARKS 1. Parametric methods for non-stationary random vibration identification and analysis were presented and their effectiveness was demonstrated. 2. The methods belong to one of the following major classes: (a) Unstructured Parameter Evolution (UPE) (b) Stochastic Parameter Evolution (SPE) (c) Deterministic Parameter Evolution (DPE) 3. The best overall performance was achieved by methods from the DPE class, followed by methods from the SPE class, and finally, the UPE class. This may be attributed to the increased structure of DPE models which leads to statistical parsimony and is in accord with the time-evolution of the actual structural dynamics. 4. Despite the progress achieved, a lot of work is necessary for addressing the numerous open issues.

NON-STATIONARY MECHANICAL VIBRATION MODELING AND ANALYSIS

NON-STATIONARY MECHANICAL VIBRATION MODELING AND ANALYSIS NON-STATIONARY MECHANICAL VIBRATION MODELING AND ANALYSIS VIA FUNCTIONAL SERIES TARMA MODELS A.G. Poulimenos and S.D. Fassois DEPARTMENT OF MECHANICAL &AERONAUTICAL ENGINEERING GR-26500 PATRAS, GREECE

More information

Multi Channel Output Only Identification of an Extendable Arm Structure Under Random Excitation: A comparison of parametric methods

Multi Channel Output Only Identification of an Extendable Arm Structure Under Random Excitation: A comparison of parametric methods Multi Channel Output Only Identification of an Extendable Arm Structure Under Random Excitation: A comparison of parametric methods Minas Spiridonakos and Spilios Fassois Stochastic Mechanical Systems

More information

Non-Stationary Random Vibration Parametric Modeling and its Application to Structural Health Monitoring

Non-Stationary Random Vibration Parametric Modeling and its Application to Structural Health Monitoring Non-Stationary Random Vibration Parametric Modeling and its Application to Structural Health Monitoring Luis David Avendaño-Valencia and Spilios D. Fassois Stochastic Mechanical Systems and Automation

More information

Stationary or Non-Stationary Random Excitation for Vibration-Based Structural Damage Detection? An exploratory study

Stationary or Non-Stationary Random Excitation for Vibration-Based Structural Damage Detection? An exploratory study Stationary or Non-Stationary Random Excitation for Vibration-Based Structural Damage Detection? An exploratory study Andriana S. GEORGANTOPOULOU & Spilios D. FASSOIS Stochastic Mechanical Systems & Automation

More information

Stationary or Non-Stationary Random Excitation for Vibration-Based Structural Damage Detection? An exploratory study

Stationary or Non-Stationary Random Excitation for Vibration-Based Structural Damage Detection? An exploratory study 6th International Symposium on NDT in Aerospace, 12-14th November 2014, Madrid, Spain - www.ndt.net/app.aerondt2014 More Info at Open Access Database www.ndt.net/?id=16938 Stationary or Non-Stationary

More information

Parametric Output Error Based Identification and Fault Detection in Structures Under Earthquake Excitation

Parametric Output Error Based Identification and Fault Detection in Structures Under Earthquake Excitation Parametric Output Error Based Identification and Fault Detection in Structures Under Earthquake Excitation J.S. Sakellariou and S.D. Fassois Department of Mechanical & Aeronautical Engr. GR 265 Patras,

More information

Parametric time-domain methods for non-stationary random vibration modelling and analysis A critical survey and comparison $

Parametric time-domain methods for non-stationary random vibration modelling and analysis A critical survey and comparison $ Mechanical Systems and Signal Processing 20 (2006) 763 816 Invited Survey Mechanical Systems and Signal Processing Parametric time-domain methods for non-stationary random vibration modelling and analysis

More information

Non-stationary functional series modeling and analysis of hardware reliability series: a comparative study using rail vehicle interfailure times

Non-stationary functional series modeling and analysis of hardware reliability series: a comparative study using rail vehicle interfailure times Reliability Engineering and System Safety 68 (2000) 169 183 www.elsevier.com/locate/ress Non-stationary functional series modeling and analysis of hardware reliability series: a comparative study using

More information

T.-C.J. Aravanis, J.S. Sakellariou and S.D. Fassois

T.-C.J. Aravanis, J.S. Sakellariou and S.D. Fassois Vibration based fault detection under variable non-measurable, operating conditions via a stochastic Functional Model method and application to railway vehicle suspensions T.-C.J. Aravanis, J.S. Sakellariou

More information

Onboard Engine FDI in Autonomous Aircraft Using Stochastic Nonlinear Modelling of Flight Signal Dependencies

Onboard Engine FDI in Autonomous Aircraft Using Stochastic Nonlinear Modelling of Flight Signal Dependencies Onboard Engine FDI in Autonomous Aircraft Using Stochastic Nonlinear Modelling of Flight Signal Dependencies Dimitrios G. Dimogianopoulos, John D. Hios and Spilios D. Fassois Stochastic Mechanical Systems

More information

Non-Stationary Time-dependent ARMA Random Vibration Modeling, Analysis & SHM with Wind Turbine Applications

Non-Stationary Time-dependent ARMA Random Vibration Modeling, Analysis & SHM with Wind Turbine Applications Non-Stationary Time-dependent ARMA Random Vibration Modeling, Analysis & SHM with Wind Turbine Applications Luis David Avendaño-Valencia Department of Mechanical Engineering and Aeronautics University

More information

the Functional Model Based Method

the Functional Model Based Method Multi-Site Damage Localization via the Functional Model Based Method Christos S. Sakaris, John S. Sakellariou and Spilios D. Fassois Stochastic Mechanical Systems & Automation (SMSA) Laboratory Department

More information

Identification of Stochastic Systems Under Multiple Operating Conditions: The Vector Dependent FP ARX Parametrization

Identification of Stochastic Systems Under Multiple Operating Conditions: The Vector Dependent FP ARX Parametrization Identification of Stochastic Systems Under Multiple Operating Conditions: The Vector Dependent FP ARX Parametrization Fotis P Kopsaftopoulos and Spilios D Fassois Abstract The problem of identifying stochastic

More information

742. Time-varying systems identification using continuous wavelet analysis of free decay response signals

742. Time-varying systems identification using continuous wavelet analysis of free decay response signals 74. Time-varying systems identification using continuous wavelet analysis of free decay response signals X. Xu, Z. Y. Shi, S. L. Long State Key Laboratory of Mechanics and Control of Mechanical Structures

More information

NONLINEAR INTEGRAL MINIMUM VARIANCE-LIKE CONTROL WITH APPLICATION TO AN AIRCRAFT SYSTEM

NONLINEAR INTEGRAL MINIMUM VARIANCE-LIKE CONTROL WITH APPLICATION TO AN AIRCRAFT SYSTEM NONLINEAR INTEGRAL MINIMUM VARIANCE-LIKE CONTROL WITH APPLICATION TO AN AIRCRAFT SYSTEM D.G. Dimogianopoulos, J.D. Hios and S.D. Fassois DEPARTMENT OF MECHANICAL & AERONAUTICAL ENGINEERING GR-26500 PATRAS,

More information

EECE Adaptive Control

EECE Adaptive Control EECE 574 - Adaptive Control Recursive Identification Algorithms Guy Dumont Department of Electrical and Computer Engineering University of British Columbia January 2012 Guy Dumont (UBC EECE) EECE 574 -

More information

9. Model Selection. statistical models. overview of model selection. information criteria. goodness-of-fit measures

9. Model Selection. statistical models. overview of model selection. information criteria. goodness-of-fit measures FE661 - Statistical Methods for Financial Engineering 9. Model Selection Jitkomut Songsiri statistical models overview of model selection information criteria goodness-of-fit measures 9-1 Statistical models

More information

Identification Methods for Structural Systems

Identification Methods for Structural Systems Prof. Dr. Eleni Chatzi Lecture 13-29 May, 2013 Courtesy of Prof. S. Fassois & Dr. F. Kopsaftopoulos, SMSA Group, University of Patras Statistical methods for SHM courtesy of Prof. S. Fassois & Dr. F. Kopsaftopoulos,

More information

Vector-dependent Functionally Pooled ARX Models for the Identification of Systems Under Multiple Operating Conditions

Vector-dependent Functionally Pooled ARX Models for the Identification of Systems Under Multiple Operating Conditions Preprints of the 16th IFAC Symposium on System Identification The International Federation of Automatic Control Vector-dependent Functionally Pooled ARX Models for the Identification of Systems Under Multiple

More information

Lecture 6: Bayesian Inference in SDE Models

Lecture 6: Bayesian Inference in SDE Models Lecture 6: Bayesian Inference in SDE Models Bayesian Filtering and Smoothing Point of View Simo Särkkä Aalto University Simo Särkkä (Aalto) Lecture 6: Bayesian Inference in SDEs 1 / 45 Contents 1 SDEs

More information

OUTPUT-ONLY STATISTICAL TIME SERIES METHODS FOR STRUCTURAL HEALTH MONITORING: A COMPARATIVE STUDY

OUTPUT-ONLY STATISTICAL TIME SERIES METHODS FOR STRUCTURAL HEALTH MONITORING: A COMPARATIVE STUDY 7th European Workshop on Structural Health Monitoring July 8-11, 2014. La Cité, Nantes, France More Info at Open Access Database www.ndt.net/?id=17198 OUTPUT-ONLY STATISTICAL TIME SERIES METHODS FOR STRUCTURAL

More information

Parameter Estimation in a Moving Horizon Perspective

Parameter Estimation in a Moving Horizon Perspective Parameter Estimation in a Moving Horizon Perspective State and Parameter Estimation in Dynamical Systems Reglerteknik, ISY, Linköpings Universitet State and Parameter Estimation in Dynamical Systems OUTLINE

More information

Location Prediction of Moving Target

Location Prediction of Moving Target Location of Moving Target Department of Radiation Oncology Stanford University AAPM 2009 Outline of Topics 1 Outline of Topics 1 2 Outline of Topics 1 2 3 Estimation of Stochastic Process Stochastic Regression:

More information

ARIMA Modelling and Forecasting

ARIMA Modelling and Forecasting ARIMA Modelling and Forecasting Economic time series often appear nonstationary, because of trends, seasonal patterns, cycles, etc. However, the differences may appear stationary. Δx t x t x t 1 (first

More information

Advanced Econometrics

Advanced Econometrics Advanced Econometrics Marco Sunder Nov 04 2010 Marco Sunder Advanced Econometrics 1/ 25 Contents 1 2 3 Marco Sunder Advanced Econometrics 2/ 25 Music Marco Sunder Advanced Econometrics 3/ 25 Music Marco

More information

Computer Intensive Methods in Mathematical Statistics

Computer Intensive Methods in Mathematical Statistics Computer Intensive Methods in Mathematical Statistics Department of mathematics johawes@kth.se Lecture 16 Advanced topics in computational statistics 18 May 2017 Computer Intensive Methods (1) Plan of

More information

The Kalman Filter ImPr Talk

The Kalman Filter ImPr Talk The Kalman Filter ImPr Talk Ged Ridgway Centre for Medical Image Computing November, 2006 Outline What is the Kalman Filter? State Space Models Kalman Filter Overview Bayesian Updating of Estimates Kalman

More information

Vibration Based Health Monitoring for a Thin Aluminum Plate: Experimental Assessment of Several Statistical Time Series Methods

Vibration Based Health Monitoring for a Thin Aluminum Plate: Experimental Assessment of Several Statistical Time Series Methods Vibration Based Health Monitoring for a Thin Aluminum Plate: Experimental Assessment of Several Statistical Time Series Methods Fotis P. Kopsaftopoulos and Spilios D. Fassois Stochastic Mechanical Systems

More information

State-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Fin. Econometrics / 53

State-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Fin. Econometrics / 53 State-space Model Eduardo Rossi University of Pavia November 2014 Rossi State-space Model Fin. Econometrics - 2014 1 / 53 Outline 1 Motivation 2 Introduction 3 The Kalman filter 4 Forecast errors 5 State

More information

Statistical Inference and Methods

Statistical Inference and Methods Department of Mathematics Imperial College London d.stephens@imperial.ac.uk http://stats.ma.ic.ac.uk/ das01/ 31st January 2006 Part VI Session 6: Filtering and Time to Event Data Session 6: Filtering and

More information

ADAPTIVE FILTER THEORY

ADAPTIVE FILTER THEORY ADAPTIVE FILTER THEORY Fourth Edition Simon Haykin Communications Research Laboratory McMaster University Hamilton, Ontario, Canada Front ice Hall PRENTICE HALL Upper Saddle River, New Jersey 07458 Preface

More information

Adaptive Filter Theory

Adaptive Filter Theory 0 Adaptive Filter heory Sung Ho Cho Hanyang University Seoul, Korea (Office) +8--0-0390 (Mobile) +8-10-541-5178 dragon@hanyang.ac.kr able of Contents 1 Wiener Filters Gradient Search by Steepest Descent

More information

Econometrics I: Univariate Time Series Econometrics (1)

Econometrics I: Univariate Time Series Econometrics (1) Econometrics I: Dipartimento di Economia Politica e Metodi Quantitativi University of Pavia Overview of the Lecture 1 st EViews Session VI: Some Theoretical Premises 2 Overview of the Lecture 1 st EViews

More information

Basis Function Selection Criterion for Modal Monitoring of Non Stationary Systems ABSTRACT RÉSUMÉ

Basis Function Selection Criterion for Modal Monitoring of Non Stationary Systems ABSTRACT RÉSUMÉ Basis Function Selection Criterion for Modal Monitoring of Non Stationary Systems Li W. 1, Vu V. H. 1, Liu Z. 1, Thomas M. 1 and Hazel B. 2 Zhaoheng.Liu@etsmtl.ca, Marc.Thomas@etsmtl.ca 1 Dynamo laboratory,

More information

Output Only Parametric Identification of a Scale Cable Stayed Bridge Structure: a comparison of vector AR and stochastic subspace methods

Output Only Parametric Identification of a Scale Cable Stayed Bridge Structure: a comparison of vector AR and stochastic subspace methods Output Only Parametric Identification of a Scale Cable Stayed Bridge Structure: a comparison of vector AR and stochastic subspace methods Fotis P. Kopsaftopoulos, Panagiotis G. Apostolellis and Spilios

More information

Time-Varying Moving Average Model for Autocovariance Nonstationary Time Series

Time-Varying Moving Average Model for Autocovariance Nonstationary Time Series Journal of Fiber Bioengineering and Informatics 7: 04) 3 6 doi: 0.3993/jfbi03040 Time-Varying Moving Average Model for Autocovariance Nonstationary Time Series Wanchun Fei a,, Lun Bai b a College of Textile

More information

Exercises - Time series analysis

Exercises - Time series analysis Descriptive analysis of a time series (1) Estimate the trend of the series of gasoline consumption in Spain using a straight line in the period from 1945 to 1995 and generate forecasts for 24 months. Compare

More information

Booth School of Business, University of Chicago Business 41914, Spring Quarter 2017, Mr. Ruey S. Tsay. Solutions to Midterm

Booth School of Business, University of Chicago Business 41914, Spring Quarter 2017, Mr. Ruey S. Tsay. Solutions to Midterm Booth School of Business, University of Chicago Business 41914, Spring Quarter 017, Mr Ruey S Tsay Solutions to Midterm Problem A: (51 points; 3 points per question) Answer briefly the following questions

More information

Lecture 2: From Linear Regression to Kalman Filter and Beyond

Lecture 2: From Linear Regression to Kalman Filter and Beyond Lecture 2: From Linear Regression to Kalman Filter and Beyond January 18, 2017 Contents 1 Batch and Recursive Estimation 2 Towards Bayesian Filtering 3 Kalman Filter and Bayesian Filtering and Smoothing

More information

Nonlinear and/or Non-normal Filtering. Jesús Fernández-Villaverde University of Pennsylvania

Nonlinear and/or Non-normal Filtering. Jesús Fernández-Villaverde University of Pennsylvania Nonlinear and/or Non-normal Filtering Jesús Fernández-Villaverde University of Pennsylvania 1 Motivation Nonlinear and/or non-gaussian filtering, smoothing, and forecasting (NLGF) problems are pervasive

More information

Time-Varying Parameters

Time-Varying Parameters Kalman Filter and state-space models: time-varying parameter models; models with unobservable variables; basic tool: Kalman filter; implementation is task-specific. y t = x t β t + e t (1) β t = µ + Fβ

More information

Using Multiple Kernel-based Regularization for Linear System Identification

Using Multiple Kernel-based Regularization for Linear System Identification Using Multiple Kernel-based Regularization for Linear System Identification What are the Structure Issues in System Identification? with coworkers; see last slide Reglerteknik, ISY, Linköpings Universitet

More information

TIME SERIES ANALYSIS. Forecasting and Control. Wiley. Fifth Edition GWILYM M. JENKINS GEORGE E. P. BOX GREGORY C. REINSEL GRETA M.

TIME SERIES ANALYSIS. Forecasting and Control. Wiley. Fifth Edition GWILYM M. JENKINS GEORGE E. P. BOX GREGORY C. REINSEL GRETA M. TIME SERIES ANALYSIS Forecasting and Control Fifth Edition GEORGE E. P. BOX GWILYM M. JENKINS GREGORY C. REINSEL GRETA M. LJUNG Wiley CONTENTS PREFACE TO THE FIFTH EDITION PREFACE TO THE FOURTH EDITION

More information

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION - Vol. V - Prediction Error Methods - Torsten Söderström

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION - Vol. V - Prediction Error Methods - Torsten Söderström PREDICTIO ERROR METHODS Torsten Söderström Department of Systems and Control, Information Technology, Uppsala University, Uppsala, Sweden Keywords: prediction error method, optimal prediction, identifiability,

More information

Prof. Dr. Roland Füss Lecture Series in Applied Econometrics Summer Term Introduction to Time Series Analysis

Prof. Dr. Roland Füss Lecture Series in Applied Econometrics Summer Term Introduction to Time Series Analysis Introduction to Time Series Analysis 1 Contents: I. Basics of Time Series Analysis... 4 I.1 Stationarity... 5 I.2 Autocorrelation Function... 9 I.3 Partial Autocorrelation Function (PACF)... 14 I.4 Transformation

More information

I. D. Landau, A. Karimi: A Course on Adaptive Control Adaptive Control. Part 9: Adaptive Control with Multiple Models and Switching

I. D. Landau, A. Karimi: A Course on Adaptive Control Adaptive Control. Part 9: Adaptive Control with Multiple Models and Switching I. D. Landau, A. Karimi: A Course on Adaptive Control - 5 1 Adaptive Control Part 9: Adaptive Control with Multiple Models and Switching I. D. Landau, A. Karimi: A Course on Adaptive Control - 5 2 Outline

More information

On Identification of Cascade Systems 1

On Identification of Cascade Systems 1 On Identification of Cascade Systems 1 Bo Wahlberg Håkan Hjalmarsson Jonas Mårtensson Automatic Control and ACCESS, School of Electrical Engineering, KTH, SE-100 44 Stockholm, Sweden. (bo.wahlberg@ee.kth.se

More information

A SARIMAX coupled modelling applied to individual load curves intraday forecasting

A SARIMAX coupled modelling applied to individual load curves intraday forecasting A SARIMAX coupled modelling applied to individual load curves intraday forecasting Frédéric Proïa Workshop EDF Institut Henri Poincaré - Paris 05 avril 2012 INRIA Bordeaux Sud-Ouest Institut de Mathématiques

More information

Financial Econometrics and Volatility Models Estimation of Stochastic Volatility Models

Financial Econometrics and Volatility Models Estimation of Stochastic Volatility Models Financial Econometrics and Volatility Models Estimation of Stochastic Volatility Models Eric Zivot April 26, 2010 Outline Likehood of SV Models Survey of Estimation Techniques for SV Models GMM Estimation

More information

ADAPTIVE FILTER THEORY

ADAPTIVE FILTER THEORY ADAPTIVE FILTER THEORY Fifth Edition Simon Haykin Communications Research Laboratory McMaster University Hamilton, Ontario, Canada International Edition contributions by Telagarapu Prabhakar Department

More information

Non-stationary Ambient Response Data Analysis for Modal Identification Using Improved Random Decrement Technique

Non-stationary Ambient Response Data Analysis for Modal Identification Using Improved Random Decrement Technique 9th International Conference on Advances in Experimental Mechanics Non-stationary Ambient Response Data Analysis for Modal Identification Using Improved Random Decrement Technique Chang-Sheng Lin and Tse-Chuan

More information

OPTIMAL FUSION OF SENSOR DATA FOR DISCRETE KALMAN FILTERING Z. G. FENG, K. L. TEO, N. U. AHMED, Y. ZHAO, AND W. Y. YAN

OPTIMAL FUSION OF SENSOR DATA FOR DISCRETE KALMAN FILTERING Z. G. FENG, K. L. TEO, N. U. AHMED, Y. ZHAO, AND W. Y. YAN Dynamic Systems and Applications 16 (2007) 393-406 OPTIMAL FUSION OF SENSOR DATA FOR DISCRETE KALMAN FILTERING Z. G. FENG, K. L. TEO, N. U. AHMED, Y. ZHAO, AND W. Y. YAN College of Mathematics and Computer

More information

TAKEHOME FINAL EXAM e iω e 2iω e iω e 2iω

TAKEHOME FINAL EXAM e iω e 2iω e iω e 2iω ECO 513 Spring 2015 TAKEHOME FINAL EXAM (1) Suppose the univariate stochastic process y is ARMA(2,2) of the following form: y t = 1.6974y t 1.9604y t 2 + ε t 1.6628ε t 1 +.9216ε t 2, (1) where ε is i.i.d.

More information

User s guide for ireg (Identification and Controller Design and Reduction) Ioan Doré Landau and Tudor-Bogdan Airimiţoaie

User s guide for ireg (Identification and Controller Design and Reduction) Ioan Doré Landau and Tudor-Bogdan Airimiţoaie User s guide for ireg (Identification and Controller Design and Reduction) Ioan Doré Landau and Tudor-Bogdan Airimiţoaie December 14, 2013 Contents 1 About ireg 3 1.1 Software requirements..........................................

More information

EECE Adaptive Control

EECE Adaptive Control EECE 574 - Adaptive Control Recursive Identification in Closed-Loop and Adaptive Control Guy Dumont Department of Electrical and Computer Engineering University of British Columbia January 2010 Guy Dumont

More information

Kalman Filter and Parameter Identification. Florian Herzog

Kalman Filter and Parameter Identification. Florian Herzog Kalman Filter and Parameter Identification Florian Herzog 2013 Continuous-time Kalman Filter In this chapter, we shall use stochastic processes with independent increments w 1 (.) and w 2 (.) at the input

More information

FOR traditional discrete-time sampled systems, the operation

FOR traditional discrete-time sampled systems, the operation 29 American Control Conference Hyatt Regency Riverfront St Louis MO USA June 1-12 29 FrA132 Least-squares based iterative parameter estimation for two-input multirate sampled-data systems Jing Lu Xinggao

More information

AN IDENTIFICATION ALGORITHM FOR ARMAX SYSTEMS

AN IDENTIFICATION ALGORITHM FOR ARMAX SYSTEMS AN IDENTIFICATION ALGORITHM FOR ARMAX SYSTEMS First the X, then the AR, finally the MA Jan C. Willems, K.U. Leuven Workshop on Observation and Estimation Ben Gurion University, July 3, 2004 p./2 Joint

More information

Statistics 910, #15 1. Kalman Filter

Statistics 910, #15 1. Kalman Filter Statistics 910, #15 1 Overview 1. Summary of Kalman filter 2. Derivations 3. ARMA likelihoods 4. Recursions for the variance Kalman Filter Summary of Kalman filter Simplifications To make the derivations

More information

Lecture 2: From Linear Regression to Kalman Filter and Beyond

Lecture 2: From Linear Regression to Kalman Filter and Beyond Lecture 2: From Linear Regression to Kalman Filter and Beyond Department of Biomedical Engineering and Computational Science Aalto University January 26, 2012 Contents 1 Batch and Recursive Estimation

More information

Online appendix to On the stability of the excess sensitivity of aggregate consumption growth in the US

Online appendix to On the stability of the excess sensitivity of aggregate consumption growth in the US Online appendix to On the stability of the excess sensitivity of aggregate consumption growth in the US Gerdie Everaert 1, Lorenzo Pozzi 2, and Ruben Schoonackers 3 1 Ghent University & SHERPPA 2 Erasmus

More information

7. Forecasting with ARIMA models

7. Forecasting with ARIMA models 7. Forecasting with ARIMA models 309 Outline: Introduction The prediction equation of an ARIMA model Interpreting the predictions Variance of the predictions Forecast updating Measuring predictability

More information

Evaluation of Some Techniques for Forecasting of Electricity Demand in Sri Lanka

Evaluation of Some Techniques for Forecasting of Electricity Demand in Sri Lanka Appeared in Sri Lankan Journal of Applied Statistics (Volume 3) 00 Evaluation of Some echniques for Forecasting of Electricity Demand in Sri Lanka.M.J. A. Cooray and M.Indralingam Department of Mathematics

More information

f-domain expression for the limit model Combine: 5.12 Approximate Modelling What can be said about H(q, θ) G(q, θ ) H(q, θ ) with

f-domain expression for the limit model Combine: 5.12 Approximate Modelling What can be said about H(q, θ) G(q, θ ) H(q, θ ) with 5.2 Approximate Modelling What can be said about if S / M, and even G / G? G(q, ) H(q, ) f-domain expression for the limit model Combine: with ε(t, ) =H(q, ) [y(t) G(q, )u(t)] y(t) =G (q)u(t) v(t) We know

More information

Lecture 4: Dynamic models

Lecture 4: Dynamic models linear s Lecture 4: s Hedibert Freitas Lopes The University of Chicago Booth School of Business 5807 South Woodlawn Avenue, Chicago, IL 60637 http://faculty.chicagobooth.edu/hedibert.lopes hlopes@chicagobooth.edu

More information

Spatially Smoothed Kernel Density Estimation via Generalized Empirical Likelihood

Spatially Smoothed Kernel Density Estimation via Generalized Empirical Likelihood Spatially Smoothed Kernel Density Estimation via Generalized Empirical Likelihood Kuangyu Wen & Ximing Wu Texas A&M University Info-Metrics Institute Conference: Recent Innovations in Info-Metrics October

More information

Optimal input design for nonlinear dynamical systems: a graph-theory approach

Optimal input design for nonlinear dynamical systems: a graph-theory approach Optimal input design for nonlinear dynamical systems: a graph-theory approach Patricio E. Valenzuela Department of Automatic Control and ACCESS Linnaeus Centre KTH Royal Institute of Technology, Stockholm,

More information

Generalized Method of Moment

Generalized Method of Moment Generalized Method of Moment CHUNG-MING KUAN Department of Finance & CRETA National Taiwan University June 16, 2010 C.-M. Kuan (Finance & CRETA, NTU Generalized Method of Moment June 16, 2010 1 / 32 Lecture

More information

Circle a single answer for each multiple choice question. Your choice should be made clearly.

Circle a single answer for each multiple choice question. Your choice should be made clearly. TEST #1 STA 4853 March 4, 215 Name: Please read the following directions. DO NOT TURN THE PAGE UNTIL INSTRUCTED TO DO SO Directions This exam is closed book and closed notes. There are 31 questions. Circle

More information

Discrete time processes

Discrete time processes Discrete time processes Predictions are difficult. Especially about the future Mark Twain. Florian Herzog 2013 Modeling observed data When we model observed (realized) data, we encounter usually the following

More information

Dynamic linear models (aka state-space models) 1

Dynamic linear models (aka state-space models) 1 Dynamic linear models (aka state-space models) 1 Advanced Econometris: Time Series Hedibert Freitas Lopes INSPER 1 Part of this lecture is based on Gamerman and Lopes (2006) Markov Chain Monte Carlo: Stochastic

More information

ANNEX A: ANALYSIS METHODOLOGIES

ANNEX A: ANALYSIS METHODOLOGIES ANNEX A: ANALYSIS METHODOLOGIES A.1 Introduction Before discussing supplemental damping devices, this annex provides a brief review of the seismic analysis methods used in the optimization algorithms considered

More information

Probabilistic Graphical Models

Probabilistic Graphical Models Probabilistic Graphical Models Brown University CSCI 2950-P, Spring 2013 Prof. Erik Sudderth Lecture 12: Gaussian Belief Propagation, State Space Models and Kalman Filters Guest Kalman Filter Lecture by

More information

AUTOMATIC CONTROL COMMUNICATION SYSTEMS LINKÖPINGS UNIVERSITET. Questions AUTOMATIC CONTROL COMMUNICATION SYSTEMS LINKÖPINGS UNIVERSITET

AUTOMATIC CONTROL COMMUNICATION SYSTEMS LINKÖPINGS UNIVERSITET. Questions AUTOMATIC CONTROL COMMUNICATION SYSTEMS LINKÖPINGS UNIVERSITET The Problem Identification of Linear and onlinear Dynamical Systems Theme : Curve Fitting Division of Automatic Control Linköping University Sweden Data from Gripen Questions How do the control surface

More information

Factor Analysis and Kalman Filtering (11/2/04)

Factor Analysis and Kalman Filtering (11/2/04) CS281A/Stat241A: Statistical Learning Theory Factor Analysis and Kalman Filtering (11/2/04) Lecturer: Michael I. Jordan Scribes: Byung-Gon Chun and Sunghoon Kim 1 Factor Analysis Factor analysis is used

More information

Introduction to time-frequency analysis. From linear to energy-based representations

Introduction to time-frequency analysis. From linear to energy-based representations Introduction to time-frequency analysis. From linear to energy-based representations Rosario Ceravolo Politecnico di Torino Dep. Structural Engineering UNIVERSITA DI TRENTO Course on «Identification and

More information

Outline 2(42) Sysid Course VT An Overview. Data from Gripen 4(42) An Introductory Example 2,530 3(42)

Outline 2(42) Sysid Course VT An Overview. Data from Gripen 4(42) An Introductory Example 2,530 3(42) Outline 2(42) Sysid Course T1 2016 An Overview. Automatic Control, SY, Linköpings Universitet An Umbrella Contribution for the aterial in the Course The classic, conventional System dentification Setup

More information

Chapter 3 - Temporal processes

Chapter 3 - Temporal processes STK4150 - Intro 1 Chapter 3 - Temporal processes Odd Kolbjørnsen and Geir Storvik January 23 2017 STK4150 - Intro 2 Temporal processes Data collected over time Past, present, future, change Temporal aspect

More information

2D Image Processing. Bayes filter implementation: Kalman filter

2D Image Processing. Bayes filter implementation: Kalman filter 2D Image Processing Bayes filter implementation: Kalman filter Prof. Didier Stricker Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche Intelligenz http://av.dfki.de

More information

Performance Analysis of an Adaptive Algorithm for DOA Estimation

Performance Analysis of an Adaptive Algorithm for DOA Estimation Performance Analysis of an Adaptive Algorithm for DOA Estimation Assimakis K. Leros and Vassilios C. Moussas Abstract This paper presents an adaptive approach to the problem of estimating the direction

More information

SINGLE DEGREE OF FREEDOM SYSTEM IDENTIFICATION USING LEAST SQUARES, SUBSPACE AND ERA-OKID IDENTIFICATION ALGORITHMS

SINGLE DEGREE OF FREEDOM SYSTEM IDENTIFICATION USING LEAST SQUARES, SUBSPACE AND ERA-OKID IDENTIFICATION ALGORITHMS 3 th World Conference on Earthquake Engineering Vancouver, B.C., Canada August -6, 24 Paper No. 278 SINGLE DEGREE OF FREEDOM SYSTEM IDENTIFICATION USING LEAST SQUARES, SUBSPACE AND ERA-OKID IDENTIFICATION

More information

TMA4285 December 2015 Time series models, solution.

TMA4285 December 2015 Time series models, solution. Norwegian University of Science and Technology Department of Mathematical Sciences Page of 5 TMA4285 December 205 Time series models, solution. Problem a) (i) The slow decay of the ACF of z t suggest that

More information

State-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Financial Econometrics / 49

State-space Model. Eduardo Rossi University of Pavia. November Rossi State-space Model Financial Econometrics / 49 State-space Model Eduardo Rossi University of Pavia November 2013 Rossi State-space Model Financial Econometrics - 2013 1 / 49 Outline 1 Introduction 2 The Kalman filter 3 Forecast errors 4 State smoothing

More information

Elements of Multivariate Time Series Analysis

Elements of Multivariate Time Series Analysis Gregory C. Reinsel Elements of Multivariate Time Series Analysis Second Edition With 14 Figures Springer Contents Preface to the Second Edition Preface to the First Edition vii ix 1. Vector Time Series

More information

3 JAA Special Publication JAA-SP-6-8E efficiency of damping estimation. It is pointed out, however, that damping is not always an appropriate index to

3 JAA Special Publication JAA-SP-6-8E efficiency of damping estimation. It is pointed out, however, that damping is not always an appropriate index to First International Symposium on Flutter and its Application, 6 3 ETENSION OF DISCRETE-TIME FLUTTER PREDICTION METHOD TO A HIGHER-MODE SYSTEM Hiroshi Torii + Meijo University, Nagoya, Japan Conventionally

More information

unit; 1m The ONJUKU COAST Total number of nodes; 600 Total nimber of elements; 1097 Onjuku Port 5 Iwawada Port No.5 No.2 No.3 No.4 No.

unit; 1m The ONJUKU COAST Total number of nodes; 600 Total nimber of elements; 1097 Onjuku Port 5 Iwawada Port No.5 No.2 No.3 No.4 No. Estimation of Tidal Currents by Kalman Filter with FEM Using Parallel Computing Naeko TAKAHASHI Abstract The purpose of this research is to estimate of tidal currents using Kalman Filter combined with

More information

Time Series Analysis. James D. Hamilton PRINCETON UNIVERSITY PRESS PRINCETON, NEW JERSEY

Time Series Analysis. James D. Hamilton PRINCETON UNIVERSITY PRESS PRINCETON, NEW JERSEY Time Series Analysis James D. Hamilton PRINCETON UNIVERSITY PRESS PRINCETON, NEW JERSEY PREFACE xiii 1 Difference Equations 1.1. First-Order Difference Equations 1 1.2. pth-order Difference Equations 7

More information

Lecture 16: State Space Model and Kalman Filter Bus 41910, Time Series Analysis, Mr. R. Tsay

Lecture 16: State Space Model and Kalman Filter Bus 41910, Time Series Analysis, Mr. R. Tsay Lecture 6: State Space Model and Kalman Filter Bus 490, Time Series Analysis, Mr R Tsay A state space model consists of two equations: S t+ F S t + Ge t+, () Z t HS t + ɛ t (2) where S t is a state vector

More information

Resampling techniques for statistical modeling

Resampling techniques for statistical modeling Resampling techniques for statistical modeling Gianluca Bontempi Département d Informatique Boulevard de Triomphe - CP 212 http://www.ulb.ac.be/di Resampling techniques p.1/33 Beyond the empirical error

More information

ECON 616: Lecture Two: Deterministic Trends, Nonstationary Processes

ECON 616: Lecture Two: Deterministic Trends, Nonstationary Processes ECON 616: Lecture Two: Deterministic Trends, Nonstationary Processes ED HERBST September 11, 2017 Background Hamilton, chapters 15-16 Trends vs Cycles A commond decomposition of macroeconomic time series

More information

A New Subspace Identification Method for Open and Closed Loop Data

A New Subspace Identification Method for Open and Closed Loop Data A New Subspace Identification Method for Open and Closed Loop Data Magnus Jansson July 2005 IR S3 SB 0524 IFAC World Congress 2005 ROYAL INSTITUTE OF TECHNOLOGY Department of Signals, Sensors & Systems

More information

On Input Design for System Identification

On Input Design for System Identification On Input Design for System Identification Input Design Using Markov Chains CHIARA BRIGHENTI Masters Degree Project Stockholm, Sweden March 2009 XR-EE-RT 2009:002 Abstract When system identification methods

More information

Topic 4 Unit Roots. Gerald P. Dwyer. February Clemson University

Topic 4 Unit Roots. Gerald P. Dwyer. February Clemson University Topic 4 Unit Roots Gerald P. Dwyer Clemson University February 2016 Outline 1 Unit Roots Introduction Trend and Difference Stationary Autocorrelations of Series That Have Deterministic or Stochastic Trends

More information

FE570 Financial Markets and Trading. Stevens Institute of Technology

FE570 Financial Markets and Trading. Stevens Institute of Technology FE570 Financial Markets and Trading Lecture 5. Linear Time Series Analysis and Its Applications (Ref. Joel Hasbrouck - Empirical Market Microstructure ) Steve Yang Stevens Institute of Technology 9/25/2012

More information

2D Image Processing. Bayes filter implementation: Kalman filter

2D Image Processing. Bayes filter implementation: Kalman filter 2D Image Processing Bayes filter implementation: Kalman filter Prof. Didier Stricker Dr. Gabriele Bleser Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche

More information

Long-Run Covariability

Long-Run Covariability Long-Run Covariability Ulrich K. Müller and Mark W. Watson Princeton University October 2016 Motivation Study the long-run covariability/relationship between economic variables great ratios, long-run Phillips

More information

Sparsity in system identification and data-driven control

Sparsity in system identification and data-driven control 1 / 40 Sparsity in system identification and data-driven control Ivan Markovsky This signal is not sparse in the "time domain" 2 / 40 But it is sparse in the "frequency domain" (it is weighted sum of six

More information

model random coefficient approach, time-dependent ARMA models, linear parameter varying ARMA models, wind turbines.

model random coefficient approach, time-dependent ARMA models, linear parameter varying ARMA models, wind turbines. Damage/Fault Diagnosis in an Operating Wind Turbine Under Uncertainty via a Vibration Response Gaussian Mixture Random Coefficient Model Based Framework Luis David Avendaño-Valencia and Spilios D. Fassois,.

More information

Problem Set 2 Solution Sketches Time Series Analysis Spring 2010

Problem Set 2 Solution Sketches Time Series Analysis Spring 2010 Problem Set 2 Solution Sketches Time Series Analysis Spring 2010 Forecasting 1. Let X and Y be two random variables such that E(X 2 ) < and E(Y 2 )

More information

Statistical Methods for Forecasting

Statistical Methods for Forecasting Statistical Methods for Forecasting BOVAS ABRAHAM University of Waterloo JOHANNES LEDOLTER University of Iowa John Wiley & Sons New York Chichester Brisbane Toronto Singapore Contents 1 INTRODUCTION AND

More information