1. Trang chủ
  2. » Khoa Học Tự Nhiên

báo cáo hóa học: " Point to point control of fractional differential linear control systems" doc

17 325 0

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 17
Dung lượng 394,26 KB

Nội dung

RESEARCH Open Access Point to point control of fractional differential linear control systems Andrzej Dzieliński * and Wiktor Malesza * Correspondence: adziel@ee.pw. edu.pl Institute of Control and Industrial Electronics Warsaw University of Technology, Koszykowa 75, 00-662 Warsaw, Poland Abstract In the article, an alternative elementary method for steering a controllable fractional linear control system with open-loop control is presented. It takes a system from an initial point to a final point in a state space, in a given finite time interval. Keywords: fractional control systems, fractional calculus, point to point control 1 Introduction Fractional integration and differentiation are generalizations of the notions of integer- order integration and differentiation. It turns out that in many real-life cases, models described by fractional differential equations much more better reflect the behavior of a phenomena than models expressed by means of the classical calculus (see, e.g., [1,2]). This idea was used successfully in various fields of science and engineering for model- ing numerous processes [3]. Mathematical fundamentals of fractional calculus are given in the monographs [4-9]. Some fractional-order controllers were developed in, e.g., [10,11]. It is also worth mentioning that there are interesting results in optimal control of fractional order systems, e.g., [12-14]. In this article, it will be shown how to steer a controllable single-input fractional lin- ear control system from a given initial state to a given final point of state space, in a given time interval. There is also shown how to derive hypothetical open-loop control functions, and some of them are presented. This method of control is an alternative to, e.g., introduced in [15], in which a derived open-loop control is based on controll- ability Gramian matrix, defined in [16] that seems to be much more complex to calcu- late than in our approach. The article is divided into two main parts: in Sect. 2 we study control systems described by the Riemann-Liouville derivatives and in Sect. 3–systems expressed by means of the Caputo derivatives. In each of these sections, we consider three cases of linear control systems: in the form of an integrator of fractional order a,intheform of sequential na-integrator, and finally, in a general (controllable) vector state space form. In Sect. 3.3, an illustrative example is given. Conclusions are given in Sect. 4. 2 Fractional control systems with Riemann-Liouville derivative Let (I α t s + g)(t ) and (D α t s + h)(t ) denote the Riemann-Liouville fractional left-sided integral and fractional derivative, respectively, of order a Î ℂ, on a finite interval of the real line [4,9]: Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 © 2011 Dzielińński and Malesza; licensee Springer. Thi s is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecomm ons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproductio n in any medium, provided the original work is prop erly cited. (I α t s + g)(t):= 1 (α) t  t s g(τ ) (t − τ ) 1−α dτ for (α) > 0, t > t s , (D α t s + h)(t):= 1 (n − α) d n dt n t  t s h(τ ) (t − τ ) α−n+1 dτ for (α) ≥ 0, t > t s , where n =[ℜ(a)] + 1, and [ℜ(a)] denotes the integer part of ℜ(a). Let us consider a fract ional-order (a Î ℝ and a > 0) differential equation of the form: (D α t s + x)(t )=f (t , x(t)), t > t s , (2:1) with the initial conditions (D α−k t s + x)(t s +) = w k , k =1, , n , (2:2) where n =[a]+1fora ∉ N,andn = a for a Î N.By (D α−k t s + x)(t s + ) ,wemeanthe following limit (D α−k t s + x)(t s +) = lim t→t s + (D α−k t s + x)(t ), k =1, , n , i.e., the limit taken in ]t s , t s + ε [for ε >0. The existence and uniqueness of solutions of (2.1) and (2.2) were considered by numerous authors, e.g., [4,8]. 2.1 Linear control system in the form of a-integrator Consider a control system of the form (D α t s + z)(t )=v(t), (2:3) where 0 <a <1,z(t) is a scalar solution of (2.3), and v(t) is a scalar control function. The aim of the control is to bring system (2.3), i.e., the state trajectory z(t), from the start point z( t s + ) = z s , (2:4) i.e., from the point z(t)=z(t s +) for t ® t s +, to the final point z( t f ) = z f , (2:5) in a finite time interval t f - t s . In other words, we are looki ng for such an open-loop control function v = v(t), which will achieve it in a finite time interval t f - t s . The start and final points will be also called the terminal points. In order to solve Equation 2.3, we need to use an initial condition of the form (D α−1 t s + z)(t s +) = (I 1−α t s + z)(t s +) = w 1 (2:6) that will correspond to condition (2.4), i.e.,wehavetofindanappropriatevaluew 1 corresponding to (2.4). To this end, initial condition (2.6) can be rewritten (see [4]) as lim t→t s + (t − t s ) 1−α z(t)= w 1  ( α ) , Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 2 of 17 from which w 1 = (α) lim t→t s + z(t) lim t→t s + (t − t s ) 1−α = z s (α) lim t→t s + (t − t s ) 1−α . (2:7) Propositio n 1. Acontrolv(t) that steers system (2.3) from the start point (2.4) to the final point (2.5) is of the form v(t)=(D α t s + ϕ)(t) , (2:8) where j(t) is an arbitrary C 1 -function satisfying ϕ ( t s ) = z s and ϕ ( t f ) = z f . (2:9) Proof. Take (2.8) as a control applied to (2.3), i.e., (D α t s + z)(t )=(D α t s + ϕ)(t) . (2:10) Integrating both sides of (2.10) by means of I α t s + , i.e., (I α t s + D α t s + z)(t )=(I α t s + D α t s + ϕ)(t) , we get (using the rule of integration given, e.g., in [4]) z(t) − (I 1 −α t s + z)(t s +)  ( α ) (t − t s ) α−1 = ϕ(t) − (I 1 −α t s + ϕ)(t s +)  ( α ) (t − t s ) α−1 . (2:11) Since j(t s )=z s , and the system starts from z(t s )=z s , we get (I 1−α t s + z)(t s +) = (I 1−α t s + ϕ)(t s +) , which finally yields z(t)=j(t). In particular, z(t f )=j(t f )=z f . □ Example 2. We want to steer system (2.3) from the start point (2.4) to the final point (2.5) by means of the control given by (2.8), where ϕ ( t ) = a 1 ( t − t s ) + a 0 , a 0 , a 1 ∈ R . (2:12) The values of coefficients a 0 and a 1 have to be chosen such that conditions (2.9) hold, i.e., from ϕ(t s )=a 0 = z s , ϕ ( t f ) = a 1 ( t f − t s ) + a 0 = z f , we calculate, for t f >t s , a 0 = z s , a 1 = z f − z s t f − t s . (2:13) Thus, polynomial (2.12) has the form ϕ(t)= z f − z s t f − t s (t − t s )+z s , and then, Equation 2.3, with control v(t)=(D α t s + ϕ)(t ) , is the following (D α t s + z)(t )=a 1 (2)  ( 2 − α ) (t − t s ) 1−α + a 0 1  ( 1 − α ) (t − t s ) −α . (2:14) Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 3 of 17 In order to show that the above-calculated control v(t) is right, we integrate (2.14) by means of I α t s + , giving z(t)− (I 1 −α t s + z)(t s +)  ( α ) (t − t s ) α−1 = a 1 (2)  ( 2 − α ) (I α t s + (t − t s ) 1−α )(t)+a 0 1  ( 1 − α ) (I α t s + (t − t s ) −α )(t) . Since the value of initial condition (I 1 −α t s + z)(t s + ) corresponding to the start point z s is given by (2.6) and (2.7), and substituting already calculated coefficients a 0 and a 1 given by (2.13), we get z(t) − z s (t − t s ) α−1 lim t→t s + (t − t s ) 1−α = z f − z s t f − t s (t − t s )+z s . (2:15) Since lim t→t s + (t − t s ) 1 −α =0 for a < 1, evaluating (2.15) at t = t s yields z(t s )=z s , and for t = t f gives z(t f )=z f . 2.2 Linear control system in the form of na-integrator Consider a control system of order na,for0<a <1,n Î N + such that na <1,given by (D n α t s + z)(t )=v(t ) (2:16) with the initial conditions (I 1−α t s + D kα t s + z)(t s +) = w k , w k ∈ R, k =0, , n − 1 , (2:17) where z(t) is a scalar solution of (2.16), (2.17), and v(t) is a scalar control function. By D k α t s + z we mean D α t s + z =D α t s + z, D kα t s + z =D α t s + D (k−1)α t s + z, k =2,3, , n . (2:18) We introduce the notion of D α t s + z (see Property 2.4 in [4]), because, in general, D α t s + D α t s + ···D α t s + z    n - t im es =D nα t s + z . Initial conditions (2.17) are equivalent (see [4]) to lim t→t s + (t − t s ) 1−α (D kα t s + z)(t )= w k  ( α ) , w k ∈ R, k =0, , n − 1 . (2:19) The aim of the control is to bring system (2.16) from the start point Z(t s ):=(z(t s ), (D α t s + z)(t s ), ,(D (n−1)α t s + z)(t s )) T =(z s0 , z s1 , , z sn−1 ) T =: Z s (2:20) at time t s , to the final point Z(t f ):=(z(t f ), (D α t s + z)(t f ), ,(D (n−1)α t s + z)(t f )) T =(z f0 , z f1 , , z fn−1 ) T =: Z f (2:21) at time t f , in the finite time interval t f - t s . For initial conditions (2.17) to correspond to the start point Z s , we cal culate (from (2.19)) Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 4 of 17 w k = (α) lim t→t s + (t − t s ) 1−α lim t→t s + (D kα t s + z)(t ) = (α) lim t→t s + (t − t s ) 1−α (D kα t s + z)(t s ) = (α) lim t→t s + (t − t s ) 1−α z sk , k =0, , n − 1 . Proposition 3. A control v(t) that s teers system (2.16) from the start point (2.20) to the final point (2.21) is of the form v(t)=(D nα t s + ϕ)(t) , where j(t) is an arbitrary C n -function satisfying D kα t s + ϕ(t s )=z sk , D kα t s + ϕ(t f )=z fk ,0≤ k ≤ n −1 , (2:22) i.e., (ϕ(t s ), ,(D (n−1)α t s + ϕ)(t s )) T = Z s and (ϕ(t f ), ,(D (n−1)α t s + ϕ)(t f )) T = Z f . For such defined conditions (2.22), the initial conditions are (I 1−α t s + D k α t s + ϕ)(t s +) = ( α) lim t→t s + (t − t s ) 1−α (D k α t s + ϕ)(t), k =0, , n − 1 . (2:23) Proof. Apply the control v(t)=D nα t s + ϕ(t ) to (2.16), and we obtain (D nα t s + z)(t )=(D nα t s + ϕ)(t) . (2:24) Next, integrating (2.24) by means of I α t s + (I α t s + D α t s + D ( n−1 ) α t s + z)(t )=(I α t s + D α t s + D ( n−1 ) α t s + ϕ)(t) , we get ( D (n−1)α t s + z)(t)− (I 1−α t s + D ( n−1 ) α t s + z)(t s +)  ( α ) (t − t s ) α−1 =( D (n−1)α t s + ϕ)(t)− (I 1−α t s + D ( n−1 ) α t s + ϕ)(t s +)  ( α ) (t − t s ) α−1 . (2:25) Since the system starts from (2.20), and (2.22) holds, i.e., D (n−1)α t s + ϕ(t s )=z sn− 1 , we get (I 1−α t s + D (n−1)α t s + z)(t s +) = (I 1−α t s + D (n−1)α t s + ϕ)(t s +) , which yields (D ( n−1 ) α t s + z)(t )=(D ( n−1 ) α t s + ϕ)(t) . (2:26) In particular, for t = t f we obtain (D ( n−1 ) α t s + z)(t f )=(D ( n−1 ) α t s + ϕ)(t f )=z fn−1 . Analogously, consecutive integrations of (2.26) by m eans of I α t s + , together for all n integrations, yields (D k α t s + z)(t s )=(D k α t s + ϕ)(t s )=z sk , k =0, , n − 1 Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 5 of 17 and (D kα t s + z)(t f )=(D kα t s + ϕ)(t f )=z fk , k =0, , n − 1 . □ One of the possible choices of function j(t)is ϕ(t)= 2n−1  i = 0 a i (I iα t s + 1)(t) , (2:27) where (I iα t s + 1)(t)= 1  ( iα +1 ) (t − t s ) iα ,0≤ i ≤ 2n −1, ((I 0 t s + 1)(t)=1 ) (2:28) satisfying (2.22). For a function of type (t - t s ) ia , the following holds (D α t s + ···D α t s +    n - t im es (t − t s ) iα )(t)=(D nα t s + (t − t s ) iα )(t)foriα +1> 0 , which is always satisfied, since we have i =0, ,2n - 1 and a >0(0<a <1).Itfol- lows that for the function (I iα t s + 1)(t ) (given by (2.28)), we have (D α t s + ···D α t s +    n - t im es I iα t s + 1)(t)=(D nα t s + I iα t s + 1)(t)=(I (i−n)α t s + 1)(t) . Thus, for the function j(t) given by (2.27), we have (D nα t s + ϕ)(t)=(D nα t s + ϕ)(t ) , and then v(t)=(D nα t s + ϕ)(t)= 2 n− 1  i = 0 a i (I (i−n)α t s + 1)(t) . Example 4. Consider control system (2.16) of order 2a (n = 2), i.e., (D 2α t s + z)(t )=v(t) , which we want to bring from the start point Z(t s ):=(z(t s ), (D α t s + z)(t s )) T =(z s0 , z s1 ) T =: Z s to the final point Z(t f ):=(z(t f ), (D α t s + z)(t f )) T =(z f0 , z f1 ) T =: Z f , in the finite time interval t f - t s . We take function j(t) in the form ϕ(t)= 3  i = 0 a i (I iα t s + 1)(t) , for which (D α t s + ϕ)(t)= 3  i = 0 a i 1 ((i −1)α +1) (t − t s ) (i−1)α . Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 6 of 17 According to (2.22), the following must be satisfied ϕ(t s )=a 0 = z s0 , (D α t s + ϕ)(t s )=a 1 = z s1 , ϕ(t f )= 3  i=0 a i (I iα t s + 1)(t f )=z f0 , (D α t s + ϕ)(t f )= 3  i = 0 a i (I (i−1)α t s + 1)(t f )=z f1 , or, in the matrix form ⎛ ⎜ ⎜ ⎝ 1000 0100 1(I α t s + 1)(t f )(I 2α t s + 1)(t f )(I 3α t s + 1)(t f ) (I −α t s + 1)(t f )1(I α t s + 1)(t f )(I 2α t s + 1)(t f ) ⎞ ⎟ ⎟ ⎠ ⎛ ⎜ ⎜ ⎝ a 0 a 1 a 2 a 3 ⎞ ⎟ ⎟ ⎠ = ⎛ ⎜ ⎜ ⎝ z s0 z s1 z f0 z f1 ⎞ ⎟ ⎟ ⎠ , (2:29) from which we can calculate coefficients a i ,0≤ i ≤ 3, assuming that t f >t s . Therefore, a control function steering the system from the start point Z s to the final point Z f ,is v(t)=(D 2α t s + ϕ)(t)= 3  i = 0 a i 1 ((i − 2)α +1) (t − t s ) (i−2)α , where a i ,0≤ i ≤ 3, are already calculated from (2.29). 2.3 Linear control system in the general state space form Consider a linear fractional control system of the form  :(D α t s + x)(t )=Ax + bu,0<α<1 , (2:30) where x(t)=(x 1 (t), , x n (t)) T Î ℝ n is a state space vector, A Î ℝ n×n , u(t) Î ℝ, b Î ℝ n×1 and (D α t s + x)(t ) = ((D α t s + x 1 )(t), ,(D α t s + x n )(t)) T . The initial conditions are (I 1−α t s + x i )(t s +) = w i , w i ∈ R,1≤ i ≤ n , or, in the equivalent form lim t→t s + (t − t s ) 1−α x i (t )= w i  ( α ) ,1≤ i ≤ n . The aim of the control is to bring the control system Λ from the start point x ( t s ) := ( x 1 ( t s ) , , x n ( t s )) T = ( x s1 , , x sn ) T =: x s (2:31) to the final point x ( t f ) := ( x 1 ( t f ) , , x n ( t f )) T = ( x f1 , , x fn ) T =: x f , (2:32) in the finite time interval t f - t s .Tothisend,sinceΛ is assumed to be controllable [15,16], i.e., rank R ( A, b ) =rank ( b, Ab, , A n−1 b ) = n, Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 7 of 17 we can change the state coordinates x to new coordinates ˜ x , in the following linear way ˜ x = Tx,whereT ∈ R n×n ,detT  = 0 such that Λ expressed in the new coordinates ˜ x = ( ˜ x 1 , , ˜ x n ) T will be in the Frobe- nius form, i.e., ˜  Fr : ˙ ˜ x = ⎛ ⎜ ⎜ ⎜ ⎜ ⎜ ⎝ 010 0 001 0 . . . . . . . . . . . . . . . 000 1 − ˜ a 0 − ˜ a 1 − ˜ a 2 − ˜ a n−1 ⎞ ⎟ ⎟ ⎟ ⎟ ⎟ ⎠ ˜ x + ⎛ ⎜ ⎜ ⎜ ⎜ ⎜ ⎝ 0 0 . . . 0 1 ⎞ ⎟ ⎟ ⎟ ⎟ ⎟ ⎠ u = ˜ A ˜ x + ˜ bu, ˜ x ∈ R n . In order to find a linear transformation T we take a row vector t 1 Î ℝ 1×n such that t 1 A j b =  00≤ j ≤ n − 2 1 j = n − 1, (2:33) which yields T = ⎛ ⎜ ⎜ ⎜ ⎝ t 1 t 1 A . . . t 1 A n−1 ⎞ ⎟ ⎟ ⎟ ⎠ . Indeed, if we take ˜ x = T x , where the first coordinate function is given by ˜ x 1 = t 1 x , and such that t 1 satisfies (2.33), then, using the linearity of Riemann-Liouville derivative, we have (D α t s + ˜ x i )(t)=t 1 A i−1 (D α t s + x)(t )=t 1 A i x = ˜ x i+1 ,1≤ i ≤ n − 1 , (D α t s + ˜ x n )(t)=t 1 A n−1 (D α t s + x)(t )=t 1 A n x + t 1 A n−1 bu = t 1 A n x + u getting ˜ x = ( t 1 , t 1 A, , t 1 A n−1 ) T x . Condition (2.33) can also be rewritten in the matrix form t 1 ( b, Ab, , A n−1 b ) = ( 0, 0, ,1 ) , which gives rise to t 1 = (0, 0, ,1)R −1 (A, b)=R − 1 ( n ) (A, b) , where R −1 ( n ) (A, b ) is the nth row of the matrix R -1 (A, b). Next, applying to the system ˜  F r a feedback of the form u( t ) = ˜ k ˜ x + v ( t ), (2:34) where ˜ k = −t 1 A n T −1 = ( ˜ a 0 , ˜ a 1 , ˜ a 2 , , ˜ a n−1 ) ∈ R 1× n and v(t) Î ℝ, we get (D α t s + ˜ x i )(t)= ˜ x i+1 ,1≤ i ≤ n − 1 , (D α t s + ˜ x n )(t)=v(t). Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 8 of 17 Denoting z = ˜ x 1 , and using notation (2.18), we get (D α t s + ˜ x i )(t)=(D iα t s + z)(t )= ˜ x i+1 ,1≤ i ≤ n − 1 , (D α t s + ˜ x n )(t)=(D nα t s + z)(t )=v(t), then (D nα t s + z)(t )=v(t) . (2:35) Since the transformation ˜ x = T x is already known, for the given start point (2.31) and final point (2.32) we can calculate corresponding terminal points expressed in the new coordinates ˜ x , i.e., ˜ x ( t s ) := ( ˜ x 1 ( t s ) , , ˜ x n ( t s )) T = Tx s = ( ˜ x s1 , , ˜ x sn ) T =: ˜ x s and ˜ x ( t f ) := ( ˜ x 1 ( t f ) , , ˜ x n ( t f )) T = Tx f = ( ˜ x f1 , , ˜ x fn ) T =: ˜ x f . Then, for system (2.35) the terminal points are the following Z(t s ):=(z(t s ), (D α t s + z)(t s ), ,(D (n−1)α t s + z)(t s )) T =( ˜ x s1 , , ˜ x sn ) T =: ˜ x s = Z s (2:36) and Z(t f ):=(z(t f ), (D α t s + z)(t f ), ,(D ( n−1 ) α t s + z)(t f )) T =( ˜ x f1 , , ˜ x fn ) T =: ˜ x f = Z f . (2:37) In such a way, we have transformed the problem of finding a control u(t) for the sys- tem (2.30) steering from the start point (2.31) to the final point (2.32), into an equiva- lent problem of finding a control v(t) for system (2.35) steering from t he start point (2.36) to the final point (2.37), which has already been explained in Sect. 2.2. To this end, we take a C n -function j(t) satisfying (2.22) for given (2.36) and (2.37). For such a function j(t), the control is v(t)=(D nα t s + ϕ)(t) . Finally, using (2.34), the desired control u(t) taking system Λ from x s to x f is the fol- lowing u (t )= ˜ k ˜ x(t)+v(t)= ˜ kTx(t)+v(t)=−R −1 ( n ) (A, b)A n x(t)+(D nα t s + ϕ)(t) . 3 Fractional control systems with Caputo derivative We will use the following definition of Caputo derivative. Let a Î ℂ and ℜ(a) ≥ 0. If a ∉ N 0 , n =[ℜ(a)] + 1, and then ( C D α t s + f )(t):= 1 (n −α) t  t s f (n) (τ ) (t − τ ) α−n+1 dτ =: (I n−α t s + D n f )(t) . If a = n Î N 0 , then ( C D n t s + f )(t)=f (n) (t ) . Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 9 of 17 Consider a differential equation, for a Î ℝ and a >0, ( C D α t s + x)(t )=f (t, x(t)), t s ≤ t ≤ t f , (3:1) with the initial conditions x (k) ( t s ) = w k , w k ∈ R, k =0, , n − 1 . (3:2) It has been already shown, e.g., in [4] that for (3.1) and (3.2) a solution exists. 3.1 Linear control system in the form of a-integrator Consider a linear fractional differential equation ( C D α t s + z)(t )=v(t), α ∈ R, α> 0 (3:3) with the initial conditions z (k) ( t s ) = w k , w k ∈ R, k =0, , n −1 , (3:4) where z(t) is a scalar solution and v(t) is a scalar control function. The aim of the control is to steer system (3.3) from the start point Z ( t s ) := ( z ( t s ) , ˙z ( t s ) , , z (n−1) ( t s )) T = ( z s0 , , z sn−1 ) T =: Z s (3:5) to the final point Z ( t f ) := ( z ( t f ) , ˙z ( t f ) , , z (n−1) ( t f )) T = ( z f0 , , z fn−1 ) T =: Z f (3:6) in a finite time interval t f - t s . In contrast to the e quation defined by means of Rie- mann-Liouville derivative, initial conditions (3.4) coincide with start point (3.5), i.e., w i = z si ,0≤ i ≤ n − 1 . Propositio n 5. Acontrolv(t) that steers system (3.3) from the start point (3.5) to the final point (3.6) is of the form v(t)=( C D α t s + ϕ)(t) , (3:7) where j(t) is an arbitrary C n -function satisfying ϕ (k) ( t s ) = z sk , ϕ (k) ( t f ) = z f k ,0≤ k ≤ n − 1 , (3:8) i.e.,  ( t s ) := ( ϕ ( t s ) , , ϕ (n−1) ( t s )) T = Z s and  ( t f ) := ( ϕ ( t f ) , , ϕ (n−1) ( t f )) T = Z f . Proof. As a control applied to (3.3) take (3.7), and then ( C D α t s + z)(t )=( C D α t s + ϕ)(t) . (3:9) Integrating (3.9) (according to the rule given by Lemma 2.22 in [4]) by means of I α t s + , i.e., (I α t s + C D α t s + z)(t )=(I α t s + C D α t s + ϕ)(t), Dzieliński and Malesza Advances in Difference Equations 2011, 2011:13 http://www.advancesindifferenceequations.com/content/2011/1/13 Page 10 of 17 [...]... model Procedings of the European Control Conference, Budapest, Hungary 2009 2 Dzieliński A, Sarwas G, Sierociuk D: Time domain validation of ultracapacitor fractional order model Procedings of the 49th IEEE Conference on Decision and Control, Atlanta, CA, USA 2010 3 Vinagre M, Monje C, Calderon A: Fractional order systems and fractional order control actions Lecture 3 of the IEEE CDC02: Fractional Calculus... 3.3 Linear control system in the general state space form Consider a controllable linear fractional control system of the form : (C Dαs + x)(t) = Ax + bu, t 0 < α < 1, where x(t) = (x1(t), , xn(t))T Î ℝn is the state space vector, A Î ℝn×n, u(t) Î ℝ, b Î ℝ n×1 and (C Dαs + x)(t) = ((C Dαs + x1 )(t), , (C Dαs + xn )(t))T The initial conditions are t t t x0 (ts ) = x0 , i i 1 ≤ i ≤ n The aim of control. .. CDC02: Fractional Calculus Applications in Automatic Control and Robotics 2002 4 Kilbas A, Srivastava H, Trujillo J: Theory and Appliccations of Fractional Differential Equations Elsevier, Amsterdam; 2006 5 Miller KS, Ross B: An Introduction to the Fractional Calculus and Fractional Differential Equations Wiley, New York; 1993 6 Oldham KB, Spanier J: The Fractional Calculus Academic Press, New York;... D: A hamiltonian formulation and a direct numerical scheme for fractional optimal control problems J Vibr Control 2007, 13(9-10):1269-1281 13 Agrawal OP, Defterli O, Baleanu D: Fractional optimal control problems with several state and control variables J Vibr Control 2010, 16(13):1967-1976 14 Baleanu D, Defterli O, Agrawal OP: A central difference numerical scheme for fractional optimal control problems... fractional differential systems Comput Eng Syst Appl 1996, 2:952-956 doi:10.1186/1687-1847-2011-13 Cite this article as: Dzieliński and Malesza: Point to point control of fractional differential linear control systems Advances in Difference Equations 2011 2011:13 Page 17 of 17 ... and (3.6) can be transformed to the following form (Dαs + y)(t) = v(t), t y(ts +) = 0, y(tf ) = zf − zs , (3:17) where y (t) = z (t) − zs (3:18) Indeed, control v(t) steering system (3.17) from the given point y(ts+) to the given point y(tf ), steers system (3.3) from the given point z(ts) to the given final point z(tf ), which follows from the inverse transformation of (3.18), i.e., z(t) = y(t) +... means that control (3.14) correctly steers the system from zs0 to zf0 Remark 7 For 0 . RESEARCH Open Access Point to point control of fractional differential linear control systems Andrzej Dzieliński * and Wiktor Malesza * Correspondence: adziel@ee.pw. edu.pl Institute of Control and Industrial Electronics. an initial point to a final point in a state space, in a given finite time interval. Keywords: fractional control systems, fractional calculus, point to point control 1 Introduction Fractional. consider three cases of linear control systems: in the form of an integrator of fractional order a,intheform of sequential na-integrator, and finally, in a general (controllable) vector state space form.

Ngày đăng: 21/06/2014, 02:20

TỪ KHÓA LIÊN QUAN

TÀI LIỆU CÙNG NGƯỜI DÙNG

TÀI LIỆU LIÊN QUAN