Time delay system identification based on optimization approaches

Size: px
Start display at page:

Download "Time delay system identification based on optimization approaches"

Transcription

1 Time delay system identification based on optimization approaches Ahlem SASSI, Saïda BEDOUI, Kamel ABDERRAHIM Numerical Control of Industrial Processes, National School of Engineers of Gabes, University of Gabes, St Omar Ibn-Khattab, 69 Gabes, Tunisia. Abstract In this paper, the problem of estimating the time delay and dynamic parameters of monovariable time delay discrete is addressed. This problem involves both the estimation of the time delay and the dynamic parameters from input-output data. In fact, we have considered our previous method which consists in minimizing a quadratic criterion using either the gradient method or the Levenberg-Marquardt method. The used criterion is deduced from a formulation allowing to define the time delay and the dynamic parameters in the same estimated vector and to build the corresponding observation vector. In order to improve the performance of this approach, we advocate the use of the quasi-newton approach based on the Broyden- Fletcher-Goldfarb-Shanno (BFGS) algorithm. Simulation results are presented to illustrate the performance of the new solution. I. INTRODUCTION This paper deals with the problem of identification of time delay systems. This problem consists in building mathematical models of processes using observed input-output data []. It has received a great attention in the last years, since the delay time is a physical phenomenon which arises in most control loops of industrial processes [],[3]. In fact, the delay may be an inherent property of the system such as in the transport processes or in the accumulation of time lags in a large number of low order systems connected in series. It may also be introduced by the time response of control loop devices such as sensors, actuators, controllers, networs, etc. The time delay identification problem involves both the estimation of the dynamic parameters and the delay time. This is one of the most difficult problems that represent an area of research where considerable wor has been done in the last years. The proposed methods for the identification of time delay systems can be classified in different ways depending on : the type of used time (continuous or discrete), the form of solution (linear or nonlinear), the type of data processing (batch or iterative or recursive), etc []-[4],[7]-[],[3],[5],[6],[8],[],[3]. Only the iterative identification approach for monovariable discrete time delay systems is considered in this paper. This approach has found applications in many fields, such as signal processing and control [],[9],[]. The early iterative method [] developed in the literature consists in using the parametrisation approach. It is based on two main steps. The first consists in using the least square algorithm to estimate the parameters of a system with a nown delay which is greater than the real delay. The second allows to deduce the delay from the first zero coefficients of the numerator. In practice, it is difficult or rather impossible to have zero coefficients from experimental /3/$3. c 3 IEEE data. Thus, a threshold must be selected. This selection leads to poor results when the input-output measurements are noisy. Another method is proposed in [6]. It is based on the use of the least square approach to identify the parameters assuming that the delay is nown. Then, it estimates the delay either by maximizing the correlation function, or by minimizing the quadratic error. This method assumes that the domain range of the delay is nown. We cite also our approach which consists in defining the time delay and the dynamic parameters in the same estimated vector and in building the corresponding observation vector [],[3]. This formulation is then used to propose a method to identify these systems by minimizing a quadratic criterion using either the gradient method or the Levenberg- Marquardt method. The gradient algorithm usually converges very slowly [5],[3],[33]. In fact, this algorithm cannot be recommended. The Levenberg-Marquardt switches between the Gauss- Newton approach and the gradient method [3],[33]. It allows us to find a solution even if the initial conditions are far to the final minimum. However, it finds only a local minimum, not a global minimum. In this paper, we propose the use the quasi- Newton approach for solved the obtained problem. The idea of quasi-newton method consists in approximating the inverse Hessian [9],[3],[33]. This represents the important advantage of the quasi-newton method since no matrix inverse has to be performed. It can be mentioned that the Broyden-Fletcher- Goldfarb-Shanno (BFGS) formula is more often adopted to approximate the inverse Hessian. This choice can be justified by the following reasons: no requirement for second order derivatives, very fast convergence and recommended for medium sizes problems ( parameters). This paper is organized as follows. Section II presents the model and its assumptions. Section III recalls the gradient and the Levenberg-Marquardt methods for the identification of time delay systems. In section IV, we propose a new solution based on the quasi-newton BFGS algorithm. Sections IV and V illustrate the performance of this method a simulation example and an experimental validation carried out on a level control process II. MODEL AND ASSUMPTIONS We consider the problem of estimating the time delay and the parameters of the following system: A(q )y()=q d B(q )u()+v() () where u() and y() are the system input and output, respectively, v() is a white noise and d is the time delay. A(q )

2 and B(q ) are two polynomials with the unit bacward shift operator q [i.e. q y()=y( )] having the orders n a and n b respectively, defined by: A(q )=+ B(q )= n a i= n b i= a i q i = +a q a na q n a b i q i = b q b nb q n b Using these expression, equation () can be rewritten as: y()= n a i= a i y( i)+ n b i= b i u( i d) () The model described above will be studied under the following assumptions: A. The polynomials A(q ) and B(q ) are coprime. A. The orders n a and n b of the model are nown. A3. The input sequence u() is a stationary ergodic process, independent of v() and is persistently exciting. A4. The disturbance v() is a sequence of independent, identically distributed random variable with zero mean and finite variance σ v A5. The input, the output and the noise are causal, i.e. u()=, y()= and v()= for. Problem statement: The goal is to develop an algorithm to estimate, simultaneously, the time delay d and the parameters (a,...,a na,b,...,b nb ). III. EXISTING METHODS Several approaches have been proposed in the litterature of the simultaneous identification of time delay and dynamic parameters of time delay systems[4]-[6],[9],[],[],[4],[6]- [8],[]-[4],[6]. These methods can be classified in different categories of solution such as the least square solution, etc. Only the optimization solution is considered in this paper. It consists in minimizing a quadratic criterion in term of the prediction error e() given by: J()= e ()= [y() ŷ()] (3) where ŷ() is the estimate of y() at iteration defined by: ŷ()= n a i= â i ( )y( i)+q ˆd( ) n b ˆb i ( )u( i) i= The estimated output ŷ() can be rewritten as: (4) ŷ()= ˆθ T ( )φ() (5) where ˆθ() the estimated parameter vector given by: ˆθ( )=[â ( ),...,â na ( ), ˆb ( ),..., ˆb nb ( )] T φ()=[ y( ),.., y( n a ),u( d),..,u( n b d)] In order to identify simultaneously the time delay and the parameters of a delay system, we have proposed to define these ones in the same estimated vector ˆθ G which we associate the corresponding observation vector φ G [],[3] such as: ˆθ G ()=[â (),...,â na (), ˆb (),..., ˆb nb (), ˆ d()] (6) φ G ()= y( ).. y( n a ) q d( ) ˆ u( ). q d( ) ˆ u( n b ) n b ˆb i= i ( )q d( ) u( i) where: u( i)= u( i) u( i ) (see appendix). We have proposed methods which consist in minimizing the quadratic criterion (3) using, in a first time the gradient search. It leads to the following iterative algorithm of computing ˆθ G () as follows: J() ˆθ G ()= ˆθ G ( ) µ() ˆθ G ( ) Deriving the quadratic criterion J() by the estimated parameter vector ˆθ( ), we obtain the following expression: J() ˆθ G ( ) = e() ŷ() e()= e() (9) ˆθ G ( ) ˆθ G ( ) Therefore, we get: (7) (8) J() ˆθ G ( ) = φ G()e() () Thus, the gradient algorithm is given by: ˆθ G ()= ˆθ G ( ) µ()φ G ()e() () where µ() is a positive scalar called step size. Then, we have proposed another method which minimizes the same criterion (3) using the Levenberg-Marquardt search: ˆθ G () = ˆθ G ( ) µ()[h()+λ I] J() ˆθ G ( ) where λ is a scalar and H() the Hessian of J() as: H() = J() ˆθ G ( ) ˆθ G ( ) T The second derivatives of the criterion (3) was approximated in [] using the small residual algorithm in [5] as: H() φ G ()φg T () () And using the expression (), the Levenberg-Marquardt algorithm is given by: ˆθ G ()= ˆθ G ( )+ µ()[φ G ()φ T G ()+λi] φ G ()e() (3) Note: If λ is too high comparing with the Hessian, this algorithm converges to the gradient one. Else, it converges to the Newton algorithm. 474

3 IV. PROPOSED APPROACHES The BFGS algorithm is one of the most efficient quasi- Newton methods for unconstrained optimization [7],[8]. It is by far the most popular quasi-newton update formula. This algorithm was proposed by Broyden, Fletcher, Goldfarb, and Shanno individually. It allows to compute ˆθ G as follows: ˆθ G ()= ˆθ G ( ) µ()[h()] J() (4) ˆθ G ( ) Using (), this expression can be rewritten as follows: ˆθ G ()= ˆθ G ( )+ µ()[h()] φ G ()e() (5) The main idea is to approximate the Hessian H() by B() as follows: B()=B( )+ W()W T () W T ()V() + B( )V()V T ()B T ( ) V T ()B( )V() (6) where B() is called the Broyden and W() is the gradient variation given by: J() ˆθ J() W()= G () ˆθ G ( ) µ() (7) V()=[B( )] φ G ()e() In order to compute the term J(), we opt for deriving J() ˆθ G () by the expression (5) of ˆθ G (). Using the series expansion of this derivation, the second order approximation is given by: J() ˆθ G () = + J() ˆθ G ( ) J() ˆθ G ( ) ˆθ G T ϕ()e()) ( )(µ()[h()] Using (7) and (), The obtained formula of W() is: W()= (ϕ()ϕt ())(µ()[h()] ϕ()e()) µ() (8) The proposed algorithm, deduced from equations (4)-(8), is summarized in algorithm. Data: u, y, θ G. N: the measurement number. choose µ(). begin For =:N Construct the observation vector φ G () using (7) Compute: V()=[H( )] ϕ()e() W()= (ϕ()ϕt ())(µ()[h()] ϕ()e()) µ() H()=H( )+ W()W T () V T ()H( )V() ˆθ G ()= ˆθ G ( )+ µ()[h()] ϕ()e() W T ()V() H( )V()V T ()H T ( ) Algorithm : BFGS Algorithm V. RESULTS We present now two examples to illustrate the effectiveness of proposed approaches. Therefore, the main purpose of these simulation examples is to compare the performance of the proposed approaches with that of gradient and Levenberg- Marquardt approaches. A. Example The simulations are performed under the following conditions: The system to be identified is persistently excited by a pseudo random binary sequence (PRBS). The estimation starts with zero initial values for the parameters and the time delay. The additive noise v() is a white noise sequence with zero mean and constant variance σ v = 4. The evaluation of the parameter error is considered to study the performance of each studied method. δ = θ ˆθ θ The simulated example is a second-order plus time delay process defined by: G(q )=q d b q + b q +a q + a q where a =., a =.36, b =.5, b =.35 and d =. The simulation results are given in Fig., Fig., Fig.3 and Fig a â b.5 ˆb a â b d 4 6 Fig.. Real and estimated evolutions of the time delay and parameters using the gradient method. 475

4 .5 a â b ˆb a â b The estimated parameters of the three methods converge to their true values The best results in term of speed of convergence are given by the proposed method. B. Example We consider a level control process. This process is composed of a set of eight identical tans illustrated in Fig.5. It is easy to remar that the output of tan (j), q j represent the input of tan (j+), q j+. Using this notation and assuming that the levels in the tans are in an operating point, we can approximate the dynamic behaviour of the level in each tan by a simple linear model: d 4 6 Fig.. Real and estimated evolutions of the time delay and parameters using the Levenberg-Marquardt method. S dh j dt = q j q j+ (9) q j+ = Kh j where S is the section of the tan and K is a constant representing the properties of the tan a â b ˆb a â b d 4 6 Fig. 3. Real and estimated evolutions of the time delay and parameters using the BFGS method..5.5 Gradient Levenberg-Marquardt BFGS Fig. 5. where Level control process. Thus, the transfer function of each tan is given by: H j (p)= K(T p+) q j(p) () T = S K Fig. 4. The parameter estimation errors δ. Based on the results presented in figures Fig., Fig., Fig.3 and Fig.4, we observe that: Then, we can deduce that the transfer function relating q (t) with the level in the tan eight h 8 (t) by: where H 8 (p)= K e (T p+) 8 q (p) () K e = f rac 476

5 The above system is a connexion in series of eight first order systems. Consequently, it can be approximated by a first order system with delay [3] defined by: K a H 8 (p)=e τ p +T a p q (p) () The delay τ, is introduced by the accumulation of time lags of the first order systems and T a and K a are the constant time and the constant of the time delay system. The discrete transfer function of () is given by: H(q )=q d b q + b q +a q (3) As a numerical example, we consider the step response of this systems with T =, K = and sampling time T e = s (Fig.6). In this case, the transfer function representing the dynamic behaviour of the global system is: H(p)= (p+) 8 (4) Fig.6 shows the presence of an important time delay. It..4.6 â ˆb Fig. 7. Real and estimated evolutions of the time delay and parameters using the gradient method â ˆb 5 Fig The step response is therefore possible to approximate the model of a very high-order complex dynamic process with a simplified model defined by (). In that way, we apply the three methods (gradient, Levenberg-Marquardt and BFGS) in order to estimate the time delay and the parameters. The obtained results are presented in Fig.7, Fig.8 and Fig.9 which show that the BFGS method gives the best results in terms of speed of convergence and precision. VI. CONCLUSION In this paper, we have considered the problem of simultaneous identification of the time delay and dynamic parameters of monovariable time delay discrete. Indeed, we have recalled our previous approach which consists in defining the time delay and the dynamic parameters in the same estimated vector and in building the corresponding observation vector. Then, this formulation is used to develop a method to identify the delay and the dynamic parameters by minimizing a quadratic criterion using either the gradient method or the Levenberg- Marquardt method. To improve the performance of this approach, we have advocated the use of the quasi-newton BFGS algorithm to solve the obtained system. Simulation and experimental examples are presented to illustrate the effectiveness of the proposed method and to compare their performance in terms of convergence speed and precision. Fig. 8. Real and estimated evolutions of the time delay and parameters using the Levenberg-Marquardt method â ˆb Fig. 9. Real and estimated evolutions of the time delay and parameters using the BFGS algorithm. APPENDIX A APPROXIMATION OF Ln(q) The shift operator and the bacward difference are given by respectively (5) and (6): 477 qu()=u(+) (5)

6 Therefore: u()=u() u( ) (6) u()=( q )u() (7) We can infer the identity between the shift operator and the bacward difference [], then: = q q = Which we apply the Logarithm function as follows: Ln(q)= Ln( ) Using the series expansion of Ln( ), the first order approximation of the shift operator is given by: Ln(q)= = q (8) ACKNOWLEDGMENT This wor was supported by the Ministry of the Higher Education and Scientific Research in Tunisia. REFERENCES [] T. Soderstrom and P. Stoica, System identification, Prentice Hall International, Series in systems and control engeneering, 989. [] J.P. Richard, Time delay systems: an overview of some recent advances and open problems, Automatica, vol. 39, pp , 3. [3] V.B. Kolmanovsii, S.I. Niculescu and K. Gu, Delay effects on stability: A survey, In proceeding of the 38th IEEE Conference on Decision and Control, vol., pp , 999. [4] X.M. Ren, A.B. Rad, P.T. Chan and W.L. Lo, Online identification of continuous-time systems with unnown time delay, IEEE Transactions on Automatic Control, vol.5, no.9, pp. 48-4, 5. [5] S.V. Draunov, W. Perruquetti, J.P. Richard and L.Beloura, Delay identification in time-delay systems using variable structure observers, Annual Reviews in Control, vol.3, no., pp.43-58, 6 [6] Y. Orlov, L. Beloura, J.P. Richard and M. Dambrine, Adaptive identification of linear time-delay systems, International Journal on Robust and non-linear Control, vol.3, no.9, pp , 3. [7] M. De la Sen, Robust adaptive control of linear time-delay systems with point time varying delays via multiestimation, Applied mathematical modelling, vol.33, no., pp , 9. [8] Q.G. Wang and Y. Zhang, Robust identification of continuous systems with dead time from step responses, Automatica, vol.37, no.3, pp ,. [9] H. Kurz and W. Goedece, Digital parameter adaptive control of process with unnown dead time, Automatica, vol.7, no., pp. 45-5, 98. [] P.J. Gawthrop and M.T. Nihtila, Identification of time delays using polynomial identification method, Systems and Control Letters, vol.5, no.4, pp. 67-7, 985. [] S.W. Sung and I.B. Lee, Prediction error identification method for continuous time processes with time delay, Industrial and Engeneering Chemistry Research, vol.4, pp ,. [] O. Gomez, Y. Orlov and I.V. Kolmanovsy, Online identification of SISO linear time-invariant delay systems from output measurements, Automatica, vol.43, no., pp , 6. [3] S. Ahmed, B. Huang and S.L. Shah, Parameter and delay estimation of continuous time models using a linear filter, Journal of Process Control, vol.43, no., pp ,. [4] A.B. Rad, W.L. Lo and K.M. Tsang, Simultaneous online identification of rational dynamics and time delay: A correlation based approach, IEEE Transactions on Control Systems Technology, vol., no.6, pp , 3. [5] W. Gao, M.L. Zhou, Y.C. Li and T. Zhang, An adaptive generalized predictive control of time varying delay system, In proceeding of the second International Conference on Machine Learning and Cybernetics, pp , 4. [6] A. Elnaggar, G.A. Dumont and A.L. Elshafei, New method for delay estimation, In proceeding of the 9th IEEE Conference on Decision and Control, vol.3, pp , 99. [7] L. Xie, Y. Liu and H. Yang, Gradient based and least squares based iterative algorithms for Matrix equations AXB+ CX T D = F, Applied Mathematics and Computation, vol.7, pp. 9-99,. [8] B. Bao, Y. Xu, J. Sheng and R. Ding, Least squares based iterative parameter estimation algorithm for multivariable controlled ARMA system modelling with finite measurement data, Mathematical and Computer Modelling, vol.53, no.9, pp ,. [9] D. Wang and F. Ding, Least squares based and gradient based iterative identification for Wiener nonlinear systems, Signal Processing, vol.9, no.5, pp. 8-89,. [] D. Wang and F. Ding, Input-output data filtering based recursive least squares identification for CARARMA systems, Digital Signal Processing, vol., no.4, pp ,. [] L. Lapidus and G.F. Pinde, Numerical Solution of Partial Diffenrential Equations in Science and Engeneering, John Wiley and sons,98. [] Bedoui. S, Ltaief. M and Abderrahim. K, New results on discrete time delay systems. International Journal of Automation and Computing, vol.9, no.6, pp , December. [3] S. Bedoui, M. Ltaief and M. Abderrahim, A new recursive algorithm for simultaneous identification of discrete time delay systems. In 6th IFAC Symposium on system identification, vol.6, no., pp. 5-57,. [4] Z. Yang, Z. Sun and C. Andersen, Nonlinear FOPDT Model Identification for the Superheat Dynamic in a Refrigeration System, Proceeding of IEEE, pp ,. [5] O. Nelles, Non linear system identification: from classical approach to neural networs and fuzzy models. Editions Springer. [6] E. Walter et L. Pronzato, Identification de modles paramtriques partir de donnes exprimentales. Editions MASSON, Paris 994. [7] QI. Chunhong, K.A. Gallivan and P.A. Absil, Recent Advances in Optimization and its Applications in Engineering. pp. 83-9,. [8] J. Nocedal and S.J. Wright, Numerical Optimization. Editions Springer 999. [9] D. Yu-Hong, Convergence properties of the BFGS algorithm. Chinese Academy of Sciences [3] J.E. Normey-Rico and E.F. Camacho, Control of Dead-time Processes Editions Springer, London 7. [3] R.O. Duda et P.E. Hart, Pattern Classification and Scene Analysis. New Yor : Wiley, 973. [3] R. Fletcher, Practical Methods of Optimization, Wiley,. [33] P.E. Gill, W. Murray, M.H. Wright: Practical Optimization, Academic Press, London

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

Optimization: Nonlinear Optimization without Constraints. Nonlinear Optimization without Constraints 1 / 23

Optimization: Nonlinear Optimization without Constraints. Nonlinear Optimization without Constraints 1 / 23 Optimization: Nonlinear Optimization without Constraints Nonlinear Optimization without Constraints 1 / 23 Nonlinear optimization without constraints Unconstrained minimization min x f(x) where f(x) is

More information

The Newton-Raphson Algorithm

The Newton-Raphson Algorithm The Newton-Raphson Algorithm David Allen University of Kentucky January 31, 2013 1 The Newton-Raphson Algorithm The Newton-Raphson algorithm, also called Newton s method, is a method for finding the minimum

More information

A Method of HVAC Process Object Identification Based on PSO

A Method of HVAC Process Object Identification Based on PSO 2017 3 45 313 doi 10.3969 j.issn.1673-7237.2017.03.004 a a b a. b. 201804 PID PID 2 TU831 A 1673-7237 2017 03-0019-05 A Method of HVAC Process Object Identification Based on PSO HOU Dan - lin a PAN Yi

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

NONLINEAR IDENTIFICATION ON BASED RBF NEURAL NETWORK

NONLINEAR IDENTIFICATION ON BASED RBF NEURAL NETWORK DAAAM INTERNATIONAL SCIENTIFIC BOOK 2011 pp. 547-554 CHAPTER 44 NONLINEAR IDENTIFICATION ON BASED RBF NEURAL NETWORK BURLAK, V. & PIVONKA, P. Abstract: This article is focused on the off-line identification

More information

EECE Adaptive Control

EECE Adaptive Control EECE 574 - Adaptive Control Basics of System Identification Guy Dumont Department of Electrical and Computer Engineering University of British Columbia January 2010 Guy Dumont (UBC) EECE574 - Basics of

More information

NON-LINEAR CONTROL OF OUTPUT PROBABILITY DENSITY FUNCTION FOR LINEAR ARMAX SYSTEMS

NON-LINEAR CONTROL OF OUTPUT PROBABILITY DENSITY FUNCTION FOR LINEAR ARMAX SYSTEMS Control 4, University of Bath, UK, September 4 ID-83 NON-LINEAR CONTROL OF OUTPUT PROBABILITY DENSITY FUNCTION FOR LINEAR ARMAX SYSTEMS H. Yue, H. Wang Control Systems Centre, University of Manchester

More information

2. Quasi-Newton methods

2. Quasi-Newton methods L. Vandenberghe EE236C (Spring 2016) 2. Quasi-Newton methods variable metric methods quasi-newton methods BFGS update limited-memory quasi-newton methods 2-1 Newton method for unconstrained minimization

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

Stochastic Quasi-Newton Methods

Stochastic Quasi-Newton Methods Stochastic Quasi-Newton Methods Donald Goldfarb Department of IEOR Columbia University UCLA Distinguished Lecture Series May 17-19, 2016 1 / 35 Outline Stochastic Approximation Stochastic Gradient Descent

More information

Methods that avoid calculating the Hessian. Nonlinear Optimization; Steepest Descent, Quasi-Newton. Steepest Descent

Methods that avoid calculating the Hessian. Nonlinear Optimization; Steepest Descent, Quasi-Newton. Steepest Descent Nonlinear Optimization Steepest Descent and Niclas Börlin Department of Computing Science Umeå University niclas.borlin@cs.umu.se A disadvantage with the Newton method is that the Hessian has to be derived

More information

Expressions for the covariance matrix of covariance data

Expressions for the covariance matrix of covariance data Expressions for the covariance matrix of covariance data Torsten Söderström Division of Systems and Control, Department of Information Technology, Uppsala University, P O Box 337, SE-7505 Uppsala, Sweden

More information

12. Prediction Error Methods (PEM)

12. Prediction Error Methods (PEM) 12. Prediction Error Methods (PEM) EE531 (Semester II, 2010) description optimal prediction Kalman filter statistical results computational aspects 12-1 Description idea: determine the model parameter

More information

Two dimensional recursive least squares for batch processes system identification

Two dimensional recursive least squares for batch processes system identification Preprints of the 10th IFAC International Symposium on Dynamics and Control of Process Systems The International Federation of Automatic Control Two dimensional recursive least squares for batch processes

More information

Lecture 7 Unconstrained nonlinear programming

Lecture 7 Unconstrained nonlinear programming Lecture 7 Unconstrained nonlinear programming Weinan E 1,2 and Tiejun Li 2 1 Department of Mathematics, Princeton University, weinan@princeton.edu 2 School of Mathematical Sciences, Peking University,

More information

Optimization 2. CS5240 Theoretical Foundations in Multimedia. Leow Wee Kheng

Optimization 2. CS5240 Theoretical Foundations in Multimedia. Leow Wee Kheng Optimization 2 CS5240 Theoretical Foundations in Multimedia Leow Wee Kheng Department of Computer Science School of Computing National University of Singapore Leow Wee Kheng (NUS) Optimization 2 1 / 38

More information

Optimization Methods

Optimization Methods Optimization Methods Decision making Examples: determining which ingredients and in what quantities to add to a mixture being made so that it will meet specifications on its composition allocating available

More information

j=1 r 1 x 1 x n. r m r j (x) r j r j (x) r j (x). r j x k

j=1 r 1 x 1 x n. r m r j (x) r j r j (x) r j (x). r j x k Maria Cameron Nonlinear Least Squares Problem The nonlinear least squares problem arises when one needs to find optimal set of parameters for a nonlinear model given a large set of data The variables x,,

More information

STAT Advanced Bayesian Inference

STAT Advanced Bayesian Inference 1 / 8 STAT 625 - Advanced Bayesian Inference Meng Li Department of Statistics March 5, 2018 Distributional approximations 2 / 8 Distributional approximations are useful for quick inferences, as starting

More information

Neural Modelling of a Yeast Fermentation Process Using Extreme Learning Machines

Neural Modelling of a Yeast Fermentation Process Using Extreme Learning Machines Neural Modelling of a Yeast Fermentation Process Using Extreme Learning Machines Maciej Ławryńczu Abstract This wor details development of dynamic neural models of a yeast fermentation chemical reactor

More information

Optimization for neural networks

Optimization for neural networks 0 - : Optimization for neural networks Prof. J.C. Kao, UCLA Optimization for neural networks We previously introduced the principle of gradient descent. Now we will discuss specific modifications we make

More information

Cramér-Rao Bounds for Estimation of Linear System Noise Covariances

Cramér-Rao Bounds for Estimation of Linear System Noise Covariances Journal of Mechanical Engineering and Automation (): 6- DOI: 593/jjmea Cramér-Rao Bounds for Estimation of Linear System oise Covariances Peter Matiso * Vladimír Havlena Czech echnical University in Prague

More information

Ch 4. Linear Models for Classification

Ch 4. Linear Models for Classification Ch 4. Linear Models for Classification Pattern Recognition and Machine Learning, C. M. Bishop, 2006. Department of Computer Science and Engineering Pohang University of Science and echnology 77 Cheongam-ro,

More information

Quasi-Newton Methods

Quasi-Newton Methods Quasi-Newton Methods Werner C. Rheinboldt These are excerpts of material relating to the boos [OR00 and [Rhe98 and of write-ups prepared for courses held at the University of Pittsburgh. Some further references

More information

Improved Damped Quasi-Newton Methods for Unconstrained Optimization

Improved Damped Quasi-Newton Methods for Unconstrained Optimization Improved Damped Quasi-Newton Methods for Unconstrained Optimization Mehiddin Al-Baali and Lucio Grandinetti August 2015 Abstract Recently, Al-Baali (2014) has extended the damped-technique in the modified

More information

Maximum Likelihood Estimation

Maximum Likelihood Estimation Maximum Likelihood Estimation Prof. C. F. Jeff Wu ISyE 8813 Section 1 Motivation What is parameter estimation? A modeler proposes a model M(θ) for explaining some observed phenomenon θ are the parameters

More information

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION - Vol. V - Recursive Algorithms - Han-Fu Chen

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION - Vol. V - Recursive Algorithms - Han-Fu Chen CONROL SYSEMS, ROBOICS, AND AUOMAION - Vol. V - Recursive Algorithms - Han-Fu Chen RECURSIVE ALGORIHMS Han-Fu Chen Institute of Systems Science, Academy of Mathematics and Systems Science, Chinese Academy

More information

Numerical Optimization Professor Horst Cerjak, Horst Bischof, Thomas Pock Mat Vis-Gra SS09

Numerical Optimization Professor Horst Cerjak, Horst Bischof, Thomas Pock Mat Vis-Gra SS09 Numerical Optimization 1 Working Horse in Computer Vision Variational Methods Shape Analysis Machine Learning Markov Random Fields Geometry Common denominator: optimization problems 2 Overview of Methods

More information

Errors-in-variables identification through covariance matching: Analysis of a colored measurement noise case

Errors-in-variables identification through covariance matching: Analysis of a colored measurement noise case 008 American Control Conference Westin Seattle Hotel Seattle Washington USA June -3 008 WeB8.4 Errors-in-variables identification through covariance matching: Analysis of a colored measurement noise case

More information

NONLINEAR PLANT IDENTIFICATION BY WAVELETS

NONLINEAR PLANT IDENTIFICATION BY WAVELETS NONLINEAR PLANT IDENTIFICATION BY WAVELETS Edison Righeto UNESP Ilha Solteira, Department of Mathematics, Av. Brasil 56, 5385000, Ilha Solteira, SP, Brazil righeto@fqm.feis.unesp.br Luiz Henrique M. Grassi

More information

Delay-Dependent Exponential Stability of Linear Systems with Fast Time-Varying Delay

Delay-Dependent Exponential Stability of Linear Systems with Fast Time-Varying Delay International Mathematical Forum, 4, 2009, no. 39, 1939-1947 Delay-Dependent Exponential Stability of Linear Systems with Fast Time-Varying Delay Le Van Hien Department of Mathematics Hanoi National University

More information

Optimal Polynomial Control for Discrete-Time Systems

Optimal Polynomial Control for Discrete-Time Systems 1 Optimal Polynomial Control for Discrete-Time Systems Prof Guy Beale Electrical and Computer Engineering Department George Mason University Fairfax, Virginia Correspondence concerning this paper should

More information

Neural Network Training

Neural Network Training Neural Network Training Sargur Srihari Topics in Network Training 0. Neural network parameters Probabilistic problem formulation Specifying the activation and error functions for Regression Binary classification

More information

Optimization II: Unconstrained Multivariable

Optimization II: Unconstrained Multivariable Optimization II: Unconstrained Multivariable CS 205A: Mathematical Methods for Robotics, Vision, and Graphics Doug James (and Justin Solomon) CS 205A: Mathematical Methods Optimization II: Unconstrained

More information

Statistics 580 Optimization Methods

Statistics 580 Optimization Methods Statistics 580 Optimization Methods Introduction Let fx be a given real-valued function on R p. The general optimization problem is to find an x ɛ R p at which fx attain a maximum or a minimum. It is of

More information

Optimal algorithm and application for point to point iterative learning control via updating reference trajectory

Optimal algorithm and application for point to point iterative learning control via updating reference trajectory 33 9 2016 9 DOI: 10.7641/CTA.2016.50970 Control Theory & Applications Vol. 33 No. 9 Sep. 2016,, (, 214122) :,.,,.,,,.. : ; ; ; ; : TP273 : A Optimal algorithm and application for point to point iterative

More information

A Sliding Mode Control based on Nonlinear Disturbance Observer for the Mobile Manipulator

A Sliding Mode Control based on Nonlinear Disturbance Observer for the Mobile Manipulator International Core Journal of Engineering Vol.3 No.6 7 ISSN: 44-895 A Sliding Mode Control based on Nonlinear Disturbance Observer for the Mobile Manipulator Yanna Si Information Engineering College Henan

More information

State Estimation by IMM Filter in the Presence of Structural Uncertainty 1

State Estimation by IMM Filter in the Presence of Structural Uncertainty 1 Recent Advances in Signal Processing and Communications Edited by Nios Mastorais World Scientific and Engineering Society (WSES) Press Greece 999 pp.8-88. State Estimation by IMM Filter in the Presence

More information

LBFGS. John Langford, Large Scale Machine Learning Class, February 5. (post presentation version)

LBFGS. John Langford, Large Scale Machine Learning Class, February 5. (post presentation version) LBFGS John Langford, Large Scale Machine Learning Class, February 5 (post presentation version) We are still doing Linear Learning Features: a vector x R n Label: y R Goal: Learn w R n such that ŷ w (x)

More information

Closed loop Identification of Four Tank Set up Using Direct Method

Closed loop Identification of Four Tank Set up Using Direct Method Closed loop Identification of Four Tan Set up Using Direct Method Mrs. Mugdha M. Salvi*, Dr.(Mrs) J. M. Nair** *(Department of Instrumentation Engg., Vidyavardhini s College of Engg. Tech., Vasai, Maharashtra,

More information

Simultaneous state and input estimation with partial information on the inputs

Simultaneous state and input estimation with partial information on the inputs Loughborough University Institutional Repository Simultaneous state and input estimation with partial information on the inputs This item was submitted to Loughborough University's Institutional Repository

More information

Constrained Optimization and Support Vector Machines

Constrained Optimization and Support Vector Machines Constrained Optimization and Support Vector Machines Man-Wai MAK Dept. of Electronic and Information Engineering, The Hong Kong Polytechnic University enmwmak@polyu.edu.hk http://www.eie.polyu.edu.hk/

More information

Maximum Likelihood Diffusive Source Localization Based on Binary Observations

Maximum Likelihood Diffusive Source Localization Based on Binary Observations Maximum Lielihood Diffusive Source Localization Based on Binary Observations Yoav Levinboo and an F. Wong Wireless Information Networing Group, University of Florida Gainesville, Florida 32611-6130, USA

More information

EL1820 Modeling of Dynamical Systems

EL1820 Modeling of Dynamical Systems EL1820 Modeling of Dynamical Systems Lecture 10 - System identification as a model building tool Experiment design Examination and prefiltering of data Model structure selection Model validation Lecture

More information

Process Identification for an SOPDT Model Using Rectangular Pulse Input

Process Identification for an SOPDT Model Using Rectangular Pulse Input Korean J. Chem. Eng., 18(5), 586-592 (2001) SHORT COMMUNICATION Process Identification for an SOPDT Model Using Rectangular Pulse Input Don Jang, Young Han Kim* and Kyu Suk Hwang Dept. of Chem. Eng., Pusan

More information

Unconstrained Multivariate Optimization

Unconstrained Multivariate Optimization Unconstrained Multivariate Optimization Multivariate optimization means optimization of a scalar function of a several variables: and has the general form: y = () min ( ) where () is a nonlinear scalar-valued

More information

Quasi-Newton Methods. Javier Peña Convex Optimization /36-725

Quasi-Newton Methods. Javier Peña Convex Optimization /36-725 Quasi-Newton Methods Javier Peña Convex Optimization 10-725/36-725 Last time: primal-dual interior-point methods Consider the problem min x subject to f(x) Ax = b h(x) 0 Assume f, h 1,..., h m are convex

More information

Prediction-based adaptive control of a class of discrete-time nonlinear systems with nonlinear growth rate

Prediction-based adaptive control of a class of discrete-time nonlinear systems with nonlinear growth rate www.scichina.com info.scichina.com www.springerlin.com Prediction-based adaptive control of a class of discrete-time nonlinear systems with nonlinear growth rate WEI Chen & CHEN ZongJi School of Automation

More information

Data Mining (Mineria de Dades)

Data Mining (Mineria de Dades) Data Mining (Mineria de Dades) Lluís A. Belanche belanche@lsi.upc.edu Soft Computing Research Group Dept. de Llenguatges i Sistemes Informàtics (Software department) Universitat Politècnica de Catalunya

More information

Local Strong Convexity of Maximum-Likelihood TDOA-Based Source Localization and Its Algorithmic Implications

Local Strong Convexity of Maximum-Likelihood TDOA-Based Source Localization and Its Algorithmic Implications Local Strong Convexity of Maximum-Likelihood TDOA-Based Source Localization and Its Algorithmic Implications Huikang Liu, Yuen-Man Pun, and Anthony Man-Cho So Dept of Syst Eng & Eng Manag, The Chinese

More information

Spectral gradient projection method for solving nonlinear monotone equations

Spectral gradient projection method for solving nonlinear monotone equations Journal of Computational and Applied Mathematics 196 (2006) 478 484 www.elsevier.com/locate/cam Spectral gradient projection method for solving nonlinear monotone equations Li Zhang, Weijun Zhou Department

More information

Linear Classifiers as Pattern Detectors

Linear Classifiers as Pattern Detectors Intelligent Systems: Reasoning and Recognition James L. Crowley ENSIMAG 2 / MoSIG M1 Second Semester 2014/2015 Lesson 16 8 April 2015 Contents Linear Classifiers as Pattern Detectors Notation...2 Linear

More information

An artificial neural networks (ANNs) model is a functional abstraction of the

An artificial neural networks (ANNs) model is a functional abstraction of the CHAPER 3 3. Introduction An artificial neural networs (ANNs) model is a functional abstraction of the biological neural structures of the central nervous system. hey are composed of many simple and highly

More information

Closed-loop Identification of Hammerstein Systems Using Iterative Instrumental Variables

Closed-loop Identification of Hammerstein Systems Using Iterative Instrumental Variables Proceedings of the 18th World Congress The International Federation of Automatic Control Closed-loop Identification of Hammerstein Systems Using Iterative Instrumental Variables Younghee Han and Raymond

More information

Second Order Sliding Mode Control for Discrete Decouplable Multivariable Systems via Input-output Models

Second Order Sliding Mode Control for Discrete Decouplable Multivariable Systems via Input-output Models International Journal of Automation and Computing 126), December 2015, 630-638 DOI: 10.1007/s11633-015-0904-z Second Order Sliding Mode Control for Discrete Decouplable Multivariable Systems via Input-output

More information

Acceleration of Levenberg-Marquardt method training of chaotic systems fuzzy modeling

Acceleration of Levenberg-Marquardt method training of chaotic systems fuzzy modeling ISSN 746-7233, England, UK World Journal of Modelling and Simulation Vol. 3 (2007) No. 4, pp. 289-298 Acceleration of Levenberg-Marquardt method training of chaotic systems fuzzy modeling Yuhui Wang, Qingxian

More information

A Discrete Robust Adaptive Iterative Learning Control for a Class of Nonlinear Systems with Unknown Control Direction

A Discrete Robust Adaptive Iterative Learning Control for a Class of Nonlinear Systems with Unknown Control Direction Proceedings of the International MultiConference of Engineers and Computer Scientists 16 Vol I, IMECS 16, March 16-18, 16, Hong Kong A Discrete Robust Adaptive Iterative Learning Control for a Class of

More information

Digital Signal Processing

Digital Signal Processing Digital Signal Processing 20 (2010) 991 999 Contents lists available at ScienceDirect Digital Signal Processing www.elsevier.com/locate/dsp Input output data filtering based recursive least squares identification

More information

Research Article Convex Polyhedron Method to Stability of Continuous Systems with Two Additive Time-Varying Delay Components

Research Article Convex Polyhedron Method to Stability of Continuous Systems with Two Additive Time-Varying Delay Components Applied Mathematics Volume 202, Article ID 689820, 3 pages doi:0.55/202/689820 Research Article Convex Polyhedron Method to Stability of Continuous Systems with Two Additive Time-Varying Delay Components

More information

Optimization II: Unconstrained Multivariable

Optimization II: Unconstrained Multivariable Optimization II: Unconstrained Multivariable CS 205A: Mathematical Methods for Robotics, Vision, and Graphics Justin Solomon CS 205A: Mathematical Methods Optimization II: Unconstrained Multivariable 1

More information

A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1

A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1 A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1 Jinglin Zhou Hong Wang, Donghua Zhou Department of Automation, Tsinghua University, Beijing 100084, P. R. China Control Systems Centre,

More information

Worst-Case Analysis of the Perceptron and Exponentiated Update Algorithms

Worst-Case Analysis of the Perceptron and Exponentiated Update Algorithms Worst-Case Analysis of the Perceptron and Exponentiated Update Algorithms Tom Bylander Division of Computer Science The University of Texas at San Antonio San Antonio, Texas 7849 bylander@cs.utsa.edu April

More information

Identification of ARX, OE, FIR models with the least squares method

Identification of ARX, OE, FIR models with the least squares method Identification of ARX, OE, FIR models with the least squares method CHEM-E7145 Advanced Process Control Methods Lecture 2 Contents Identification of ARX model with the least squares minimizing the equation

More information

Trust Regions. Charles J. Geyer. March 27, 2013

Trust Regions. Charles J. Geyer. March 27, 2013 Trust Regions Charles J. Geyer March 27, 2013 1 Trust Region Theory We follow Nocedal and Wright (1999, Chapter 4), using their notation. Fletcher (1987, Section 5.1) discusses the same algorithm, but

More information

Improving L-BFGS Initialization for Trust-Region Methods in Deep Learning

Improving L-BFGS Initialization for Trust-Region Methods in Deep Learning Improving L-BFGS Initialization for Trust-Region Methods in Deep Learning Jacob Rafati http://rafati.net jrafatiheravi@ucmerced.edu Ph.D. Candidate, Electrical Engineering and Computer Science University

More information

EECE Adaptive Control

EECE Adaptive Control EECE 574 - Adaptive Control Overview Guy Dumont Department of Electrical and Computer Engineering University of British Columbia Lectures: Thursday 09h00-12h00 Location: PPC 101 Guy Dumont (UBC) EECE 574

More information

Refined Instrumental Variable Methods for Identifying Hammerstein Models Operating in Closed Loop

Refined Instrumental Variable Methods for Identifying Hammerstein Models Operating in Closed Loop Refined Instrumental Variable Methods for Identifying Hammerstein Models Operating in Closed Loop V. Laurain, M. Gilson, H. Garnier Abstract This article presents an instrumental variable method dedicated

More information

Adaptive Predictive Observer Design for Class of Uncertain Nonlinear Systems with Bounded Disturbance

Adaptive Predictive Observer Design for Class of Uncertain Nonlinear Systems with Bounded Disturbance International Journal of Control Science and Engineering 2018, 8(2): 31-35 DOI: 10.5923/j.control.20180802.01 Adaptive Predictive Observer Design for Class of Saeed Kashefi *, Majid Hajatipor Faculty of

More information

Higher-Order Methods

Higher-Order Methods Higher-Order Methods Stephen J. Wright 1 2 Computer Sciences Department, University of Wisconsin-Madison. PCMI, July 2016 Stephen Wright (UW-Madison) Higher-Order Methods PCMI, July 2016 1 / 25 Smooth

More information

MATH 4211/6211 Optimization Quasi-Newton Method

MATH 4211/6211 Optimization Quasi-Newton Method MATH 4211/6211 Optimization Quasi-Newton Method Xiaojing Ye Department of Mathematics & Statistics Georgia State University Xiaojing Ye, Math & Stat, Georgia State University 0 Quasi-Newton Method Motivation:

More information

Research Article Weighted Measurement Fusion White Noise Deconvolution Filter with Correlated Noise for Multisensor Stochastic Systems

Research Article Weighted Measurement Fusion White Noise Deconvolution Filter with Correlated Noise for Multisensor Stochastic Systems Mathematical Problems in Engineering Volume 2012, Article ID 257619, 16 pages doi:10.1155/2012/257619 Research Article Weighted Measurement Fusion White Noise Deconvolution Filter with Correlated Noise

More information

Iterative Controller Tuning Using Bode s Integrals

Iterative Controller Tuning Using Bode s Integrals Iterative Controller Tuning Using Bode s Integrals A. Karimi, D. Garcia and R. Longchamp Laboratoire d automatique, École Polytechnique Fédérale de Lausanne (EPFL), 05 Lausanne, Switzerland. email: alireza.karimi@epfl.ch

More information

Exploring Granger Causality for Time series via Wald Test on Estimated Models with Guaranteed Stability

Exploring Granger Causality for Time series via Wald Test on Estimated Models with Guaranteed Stability Exploring Granger Causality for Time series via Wald Test on Estimated Models with Guaranteed Stability Nuntanut Raksasri Jitkomut Songsiri Department of Electrical Engineering, Faculty of Engineering,

More information

UTILIZING PRIOR KNOWLEDGE IN ROBUST OPTIMAL EXPERIMENT DESIGN. EE & CS, The University of Newcastle, Australia EE, Technion, Israel.

UTILIZING PRIOR KNOWLEDGE IN ROBUST OPTIMAL EXPERIMENT DESIGN. EE & CS, The University of Newcastle, Australia EE, Technion, Israel. UTILIZING PRIOR KNOWLEDGE IN ROBUST OPTIMAL EXPERIMENT DESIGN Graham C. Goodwin James S. Welsh Arie Feuer Milan Depich EE & CS, The University of Newcastle, Australia 38. EE, Technion, Israel. Abstract:

More information

Model Identification and Validation for a Heating System using MATLAB System Identification Toolbox

Model Identification and Validation for a Heating System using MATLAB System Identification Toolbox IOP Conference Series: Materials Science and Engineering OPEN ACCESS Model Identification and Validation for a Heating System using MATLAB System Identification Toolbox To cite this article: Muhammad Junaid

More information

Improving performance and stability of MRI methods in closed-loop

Improving performance and stability of MRI methods in closed-loop Preprints of the 8th IFAC Symposium on Advanced Control of Chemical Processes The International Federation of Automatic Control Improving performance and stability of MRI methods in closed-loop Alain Segundo

More information

Optimization Methods

Optimization Methods Optimization Methods Categorization of Optimization Problems Continuous Optimization Discrete Optimization Combinatorial Optimization Variational Optimization Common Optimization Concepts in Computer Vision

More information

Nonlinear Controller Design of the Inverted Pendulum System based on Extended State Observer Limin Du, Fucheng Cao

Nonlinear Controller Design of the Inverted Pendulum System based on Extended State Observer Limin Du, Fucheng Cao International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 015) Nonlinear Controller Design of the Inverted Pendulum System based on Extended State Observer Limin Du,

More information

RECURSIVE SUBSPACE IDENTIFICATION IN THE LEAST SQUARES FRAMEWORK

RECURSIVE SUBSPACE IDENTIFICATION IN THE LEAST SQUARES FRAMEWORK RECURSIVE SUBSPACE IDENTIFICATION IN THE LEAST SQUARES FRAMEWORK TRNKA PAVEL AND HAVLENA VLADIMÍR Dept of Control Engineering, Czech Technical University, Technická 2, 166 27 Praha, Czech Republic mail:

More information

Machine Learning 2017

Machine Learning 2017 Machine Learning 2017 Volker Roth Department of Mathematics & Computer Science University of Basel 21st March 2017 Volker Roth (University of Basel) Machine Learning 2017 21st March 2017 1 / 41 Section

More information

Quadratic Extended Filtering in Nonlinear Systems with Uncertain Observations

Quadratic Extended Filtering in Nonlinear Systems with Uncertain Observations Applied Mathematical Sciences, Vol. 8, 2014, no. 4, 157-172 HIKARI Ltd, www.m-hiari.com http://dx.doi.org/10.12988/ams.2014.311636 Quadratic Extended Filtering in Nonlinear Systems with Uncertain Observations

More information

Lazy learning for control design

Lazy learning for control design Lazy learning for control design Gianluca Bontempi, Mauro Birattari, Hugues Bersini Iridia - CP 94/6 Université Libre de Bruxelles 5 Bruxelles - Belgium email: {gbonte, mbiro, bersini}@ulb.ac.be Abstract.

More information

Pattern Classification

Pattern Classification Pattern Classification All materials in these slides were taen from Pattern Classification (2nd ed) by R. O. Duda,, P. E. Hart and D. G. Stor, John Wiley & Sons, 2000 with the permission of the authors

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

A Systematic Design of Emulators for Multivariable Non Square and Nonlinear Systems

A Systematic Design of Emulators for Multivariable Non Square and Nonlinear Systems International Journal of Automation and Computing 14(6), December 2017, 742-754 DOI: 10.1007/s11633-016-0963-9 A Systematic Design of Emulators for Multivariable Non Square and Nonlinear Systems Nesrine

More information

Open Access Permanent Magnet Synchronous Motor Vector Control Based on Weighted Integral Gain of Sliding Mode Variable Structure

Open Access Permanent Magnet Synchronous Motor Vector Control Based on Weighted Integral Gain of Sliding Mode Variable Structure Send Orders for Reprints to reprints@benthamscienceae The Open Automation and Control Systems Journal, 5, 7, 33-33 33 Open Access Permanent Magnet Synchronous Motor Vector Control Based on Weighted Integral

More information

Closed-Loop Identification using Routine Operating Data: the Effect of Time Delay

Closed-Loop Identification using Routine Operating Data: the Effect of Time Delay Closed-Loop Identification using Routine Operating Data: the Effect of Time Delay Yuri A.W. Shardt*, Biao Huang* *University of Alberta, Edmonton, Alberta, Canada, T6G 2V4 (Tel: 780-492-906; e-mail: {yuri.shardt,

More information

THE RELATIONSHIPS BETWEEN CG, BFGS, AND TWO LIMITED-MEMORY ALGORITHMS

THE RELATIONSHIPS BETWEEN CG, BFGS, AND TWO LIMITED-MEMORY ALGORITHMS Furman University Electronic Journal of Undergraduate Mathematics Volume 12, 5 20, 2007 HE RELAIONSHIPS BEWEEN CG, BFGS, AND WO LIMIED-MEMORY ALGORIHMS ZHIWEI (ONY) QIN Abstract. For the solution of linear

More information

Stability Analysis of Linear Systems with Time-varying State and Measurement Delays

Stability Analysis of Linear Systems with Time-varying State and Measurement Delays Proceeding of the th World Congress on Intelligent Control and Automation Shenyang, China, June 29 - July 4 24 Stability Analysis of Linear Systems with ime-varying State and Measurement Delays Liang Lu

More information

Reading Group on Deep Learning Session 1

Reading Group on Deep Learning Session 1 Reading Group on Deep Learning Session 1 Stephane Lathuiliere & Pablo Mesejo 2 June 2016 1/31 Contents Introduction to Artificial Neural Networks to understand, and to be able to efficiently use, the popular

More information

Norm invariant discretization for sampled-data fault detection

Norm invariant discretization for sampled-data fault detection Automatica 41 (25 1633 1637 www.elsevier.com/locate/automatica Technical communique Norm invariant discretization for sampled-data fault detection Iman Izadi, Tongwen Chen, Qing Zhao Department of Electrical

More information

Static temperature analysis and compensation of MEMS gyroscopes

Static temperature analysis and compensation of MEMS gyroscopes Int. J. Metrol. Qual. Eng. 4, 209 214 (2013) c EDP Sciences 2014 DOI: 10.1051/ijmqe/2013059 Static temperature analysis and compensation of MEMS gyroscopes Q.J. Tang 1,2,X.J.Wang 1,Q.P.Yang 2,andC.Z.Liu

More information

Quasi-Newton Methods. Zico Kolter (notes by Ryan Tibshirani, Javier Peña, Zico Kolter) Convex Optimization

Quasi-Newton Methods. Zico Kolter (notes by Ryan Tibshirani, Javier Peña, Zico Kolter) Convex Optimization Quasi-Newton Methods Zico Kolter (notes by Ryan Tibshirani, Javier Peña, Zico Kolter) Convex Optimization 10-725 Last time: primal-dual interior-point methods Given the problem min x f(x) subject to h(x)

More information

Extremum Seeking for Dead-Zone Compensation and Its Application to a Two-Wheeled Robot

Extremum Seeking for Dead-Zone Compensation and Its Application to a Two-Wheeled Robot Extremum Seeking for Dead-Zone Compensation and Its Application to a Two-Wheeled Robot Dessy Novita Graduate School of Natural Science and Technology, Kanazawa University, Kakuma, Kanazawa, Ishikawa, Japan

More information

Model Based Fault Detection and Diagnosis Using Structured Residual Approach in a Multi-Input Multi-Output System

Model Based Fault Detection and Diagnosis Using Structured Residual Approach in a Multi-Input Multi-Output System SERBIAN JOURNAL OF ELECTRICAL ENGINEERING Vol. 4, No. 2, November 2007, 133-145 Model Based Fault Detection and Diagnosis Using Structured Residual Approach in a Multi-Input Multi-Output System A. Asokan

More information

Chapter 4. Unconstrained optimization

Chapter 4. Unconstrained optimization Chapter 4. Unconstrained optimization Version: 28-10-2012 Material: (for details see) Chapter 11 in [FKS] (pp.251-276) A reference e.g. L.11.2 refers to the corresponding Lemma in the book [FKS] PDF-file

More information

Pattern Recognition and Machine Learning. Bishop Chapter 6: Kernel Methods

Pattern Recognition and Machine Learning. Bishop Chapter 6: Kernel Methods Pattern Recognition and Machine Learning Chapter 6: Kernel Methods Vasil Khalidov Alex Kläser December 13, 2007 Training Data: Keep or Discard? Parametric methods (linear/nonlinear) so far: learn parameter

More information

KNOWN approaches for improving the performance of

KNOWN approaches for improving the performance of IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II: EXPRESS BRIEFS, VOL. 58, NO. 8, AUGUST 2011 537 Robust Quasi-Newton Adaptive Filtering Algorithms Md. Zulfiquar Ali Bhotto, Student Member, IEEE, and Andreas

More information

Automated Tuning of the Nonlinear Complementary Filter for an Attitude Heading Reference Observer

Automated Tuning of the Nonlinear Complementary Filter for an Attitude Heading Reference Observer Automated Tuning of the Nonlinear Complementary Filter for an Attitude Heading Reference Observer Oscar De Silva, George K.I. Mann and Raymond G. Gosine Faculty of Engineering and Applied Sciences, Memorial

More information