You are on page 1of 6

Advanced Studies in Theoretical Physics

Vol. 11, 2017, no. 12, 725 - 730


On a Periodic Problem Using

Variational Calculus
Pedro Pablo Cárdenas Alzate

Department of Mathematics and GEDNOL

Universidad Tecnológica de Pereira
Pereira, Colombia

Fernando Mesa

Department of Mathematics and GEDNOL

Universidad Tecnológica de Pereira
Pereira, Colombia

Carlos Alberto Rodrı́guez Varela

Department of Mathematics
Universidad Tecnológica de Pereira
Pereira, Colombia

c 2017 Pedro Pablo Cárdenas Alzate et al. This article is distributed under
the Creative Commons Attribution License, which permits unrestricted use, distribution,
and reproduction in any medium, provided the original work is properly cited.


In this work we solve a periodic problem for a continuously differen-

tiable function. Similarly, we prove that the function φ associated with
the problem and defined in HT1 reaches a minimum using the classical
theorems of the calculation of variations.

Keywords: Periodic problem, variational calculus, minimum.

726 Pedro Pablo Cárdenas Alzate et al.

1 Introduction
Minimization principles form one of the most wide-ranging means of formulat-
ing mathematical models governing physical systems. In this paper, we will
develop the basic mathematical analysis of nonlinear minimization principles
on infinite-dimensional function spaces a subject known as the calculus of vari-
ations, for reasons that will be explained as soon as we present the basic ideas.
Classical solutions to minimization problems in the calculus of variations are
prescribed by boundary value problems involving certain types of differential
equations, known as the associated Euler-Lagrange equations.

From a variational point of view it is often difficult to distinguish solutions

with collisions from classical solutions, since generally the presence of collisions
does not result in a significant increment in the value of the action functional.
Minimization problems that can be analyzed by the calculus of variations (vari-
ational calculus) serve to characterize the equilibrium configurations of almost
all continuous physical systems. Many geometrical configurations, such as min-
imal surfaces, can be conveniently formulated as optimization problems. More-
over, numerical approximations to the equilibrium solutions of such boundary
value problems are based on a nonlinear finite element approach that reduced
the infinite-dimensional minimization problem to a finite-dimensional problem,
to which we can apply the optimization techniques of variational calculus.

Let φ(x) : HT1 → R be a function defined by

Z T 0 2
|x (t)|
φ(x) = + F (t, x(t)) dt.
0 2
Here, F : [0, T ] × RN → R is a function continuous such that:
• Let x 7→ F (t, x) be continuously differentiable for all t ∈ [0, T ]; we denote
∂F ∂F
∇F (t, x) = (t, x), · · · , (t, x)
∂x1 ∂xN

• There is a ∈ C([0, ∞[, [0, ∞[), b ∈ C[0, T ] such that

|F (t, x)| ≤ a(|x|)b(t), |∇F | ≤ a(|x|)b(t), ∀(t, x) ∈ [0, T ] × RN .

2 Solution of periodic problem

In this section we will show that there is x ∈ HT1 such that φ0 (x) = 0 and
therefore the periodic problem
x00 (t) = ∇F (t, x(t)), t ∈ [0, T ]
On a periodic problem using variational calculus 727

x(0) = x(T ), x0 (0) = x0 (T )

has a solution. In fact, first we can associate with this problem the function
given by

L(t, x, x0 ) = |x0 |2 + F (t, x), (1)
with L : [0, T ] × RN × RN → R; F : [0, T ] × RN → R and (t, x, x0 ) ∈ [0, T ] ×
RN × RN . Now, we can see that p → L(t, x, p) is a convex function for every
(t, x) ∈ [0, T ] × RN . Therefore, let

Ω : {x(t) ∈ C([0, T ], RN ) | x(0) = x(T ), x0 (0) = x0 (T )},

and Z T
A(x) : L(t, x(t), x0 (t)) dt.

Therefore, we know (by the classical theorems of variational calculus) that

the periodic problem is a necessary condition for the variational problem
min A(x), A(x) = L(t, x(t), x0 (t)) dt, (2)
x∈Ω 0

i.e., if we obtain a solution of (2), then we obtain a solution of the periodic

problem. Now, for a continuously differentiable function, let’s say

S : [0, T ] × RN → R,

which satisfies
∂S ∂S
S(0, x) = S(T, x), (0, x) = (T, x), x ∈ RN (3)
∂t ∂t
so we define
L̃(t, x, p) = L(t, x, p) − (t, x) − h∇S(t, x), pi

Ã(x) = L̃(t, x(t), x0 (t)) dt,

then, for any x ∈ Ω, we have that

728 Pedro Pablo Cárdenas Alzate et al.

A(x) − Ã(x) = L(t, x(t), x (t)) dt − L̃(t, x(t), x0 (t)) dt
0 0
Z T  
= (t, x(t)) + h∇S(t, x(t)), x0 (t)i dt
0 ∂t
= (d/dt)S(t, x(t)) dt = S(T, x(T )) − S(0, x(0)) = 0.

Therefore, the variational problem (2) is equivalent to the problem

min Ã(x), Ã(x) = L̃(t, x(t), x0 (t)) dt. (4)
x∈Ω 0

We can conclude that we can obtain the solution of the original problem
by solving (4).

Now, we look for a function S that satisfies the following conditions:

1. For any (t, x, p) ∈ [0, T ] × RN → RN ,

L̃(t, x, p) ≥ 0. (5)

2. For any (t, x) ∈ [0, T ] × RN , the equation

L̃(t, x, p) = 0. (6)

has a solution z = z(t, x) which satisfies

z(0, x) = z(T, x), (7)

i.e., we look for a function S which satisfies (3) and such that the function L̃ is
not negative, with the property only for each (t, x) ∈ [0, T ] × RN , this function
L̃ reaches a minimum in z(t, x).

Finally, we must prove the following result:

Let S : [0, T ]×RN → R a continuously differentiable function that satisfies

(3). If x∗ : [0, T ] → RN is a solution of the following problem

x0 (t) = z(t, x(t)), t ∈ [0, T ]; x(0) = x(T ), (8)

On a periodic problem using variational calculus 729

then x∗ is a minimum of the problem (2), therefore x∗ is a periodic solution of

the original problem.

To prove it, let x∗ : [0, T ] → RN a solution of (8), then (7) and x(0) = x(T )
0 0
given x∗ (0) = x∗ (T ), which implies that x∗ ∈ Ω. Using (5) and (6) we have

L̃(t, x(t), x0 (t)) ≥ L̃(t, x∗ (t), z(t, x∗ (t))) = 0.

Therefore, for any x ∈ Ω we have

∗ 0
A(x) − A(x ) = L(t, x(t), x0 (t)) − L(t, x∗ (t), x∗ (t)) dt
= L̃(t, x(t), x0 (t)) − L̃(t, x∗ (t), z(t, x∗ (t))) dt
Z T   
∂S ∗ ∗ ∗
− (t, x (t)) + h∇S(t, x (t)), z(t, x (t))i dt
0 ∂t
= L̃(t, x(t), x0 (t)) − L̃(t, x∗ (t), z(t, x∗ (t))) dt

− (d/dt)S(t, x (t)) dt + (d/dt)S(t, x(t)) dt
0 0
= L̃(t, x(t), x0 (t)) − L̃(t, x∗ (t), z(t, x∗ (t))) dt
− (S(T, x∗ (T )) − S(0, x∗ (0))) + (S(T, x(T )) − S(0, x(0)))
= L̃(t, x(t), x0 (t)) − L̃(t, x∗ (t), z(t, x∗ (t))) dt ≥ 0,

i.e., we obtain
A(x) ≥ A(x∗ ), ∀x ∈ Ω.

3 Conclusion
In this paper we have tried some classic results of the calculation of variations
applied to the existence of a periodic problem. We have shown the existence
of a minimum for the associated periodic problem.

Acknowledgements. We would like to thank the referee for his valuable

suggestions that improved the presentation of this paper and our gratitude
730 Pedro Pablo Cárdenas Alzate et al.

to the Department of Mathematics of the Universidad Tecnológica de Pereira

(Colombia) and the group GEDNOL.

[1] W. Yang, Applied Numerical Methods Using MatLab, John Wiley, 2005.

[2] O. Agrawal, Generalized Euler-Lagrange equations and transversality con-

ditions for FVPs in terms of the Caputo derivative, Journal of Vibration
and Control, 13 (2007), 1217-1237.

[3] Dinh Nho Hao, B.T. Johansson, D. Lesnic, Pham Minh Hien, A varia-
tional method and approximations of a Cauchy problem for elliptic equa-
tions, J. Algorithms Comput. Technol., 4 (2010), 89-119.

[4] M. Fontelos, Fundamentos Matemáticos de la Ingenierı́a, Vol. 1, Librerı́a-

Editorial Dykinson, 2007.

[5] K. Atkinson, The Numerical Solution of Integral Equations of the Second

Kind, Cambridge University Press,Cambridge, 1997.

[6] P.P. Cárdenas Alzate, A Survey of the Implementation of Numerical

Schemes for Linear Advection Equation, Advances in Pure Mathematics,
4 (2014), 467-479.

[7] R. El-Nabulsi, A periodic functional approach to the calculus of variations

and the problem of time-dependent damped harmonic oscillators, Applied
Mathematics Letters, 2 (2011), 1647-1653.

[8] A. Doucet, N. de Freitas, N. Gordon, An Introduction to Sequential Monte

Carlo Methods, Sequential Monte Carlo Methods in Practice, Springer
2001, 3-14. 1

Received: November 30, 2017; December 19, 2017