Research article

Applying fixed point techniques to solve fractional differential inclusions under new boundary conditions

  • Received: 02 February 2024 Revised: 16 April 2024 Accepted: 19 April 2024 Published: 29 April 2024
  • MSC : 34A05, 34B15, 47H10

  • Many scholars have lately explored fractional-order boundary value issues with a variety of conditions, including classical, nonlocal, multipoint, periodic/anti-periodic, fractional-order, and integral boundary conditions. In this manuscript, the existence and uniqueness of solutions to sequential fractional differential inclusions via a novel set of nonlocal boundary conditions were investigated. The existence results were presented under a new class of nonlocal boundary conditions, Carathéodory functions, and Lipschitz mappings. Further, fixed-point techniques have been applied to study the existence of results under convex and non-convex multi-valued mappings. Ultimately, to support our findings, we analyzed an illustrative example.

    Citation: Murugesan Manigandan, Kannan Manikandan, Hasanen A. Hammad, Manuel De la Sen. Applying fixed point techniques to solve fractional differential inclusions under new boundary conditions[J]. AIMS Mathematics, 2024, 9(6): 15505-15542. doi: 10.3934/math.2024750

    Related Papers:

    [1] Huizhang Yang, Wei Liu, Yunmei Zhao . Lie symmetry reductions and exact solutions to a generalized two-component Hunter-Saxton system. AIMS Mathematics, 2021, 6(2): 1087-1100. doi: 10.3934/math.2021065
    [2] Nouf Almutiben, Edward L. Boone, Ryad Ghanam, G. Thompson . Classification of the symmetry Lie algebras for six-dimensional co-dimension two Abelian nilradical Lie algebras. AIMS Mathematics, 2024, 9(1): 1969-1996. doi: 10.3934/math.2024098
    [3] Nouf Almutiben, Ryad Ghanam, G. Thompson, Edward L. Boone . Symmetry analysis of the canonical connection on Lie groups: six-dimensional case with abelian nilradical and one-dimensional center. AIMS Mathematics, 2024, 9(6): 14504-14524. doi: 10.3934/math.2024705
    [4] Rutwig Campoamor-Stursberg, Eduardo Fernández-Saiz, Francisco J. Herranz . Generalized Buchdahl equations as Lie–Hamilton systems from the "book" and oscillator algebras: quantum deformations and their general solution. AIMS Mathematics, 2025, 10(3): 6873-6909. doi: 10.3934/math.2025315
    [5] Baiying He, Siyu Gao . The nonisospectral integrable hierarchies of three generalized Lie algebras. AIMS Mathematics, 2024, 9(10): 27361-27387. doi: 10.3934/math.20241329
    [6] Yang Pan, Yanyong Hong . Varieties of a class of elementary subalgebras. AIMS Mathematics, 2022, 7(2): 2084-2101. doi: 10.3934/math.2022119
    [7] Zhongxian Huang . Biderivations of the extended Schrödinger-Virasoro Lie algebra. AIMS Mathematics, 2023, 8(12): 28808-28817. doi: 10.3934/math.20231476
    [8] Yuqiang Feng, Jicheng Yu . Lie symmetry analysis of fractional ordinary differential equation with neutral delay. AIMS Mathematics, 2021, 6(4): 3592-3605. doi: 10.3934/math.2021214
    [9] Amjad Hussain, Muhammad Khubaib Zia, Kottakkaran Sooppy Nisar, Velusamy Vijayakumar, Ilyas Khan . Lie analysis, conserved vectors, nonlinear self-adjoint classification and exact solutions of generalized (N+1)-dimensional nonlinear Boussinesq equation. AIMS Mathematics, 2022, 7(7): 13139-13168. doi: 10.3934/math.2022725
    [10] Mauro Costantini, Pierpaolo Soravia . On the optimal second order decrease rate for nonlinear and symmetric control systems. AIMS Mathematics, 2024, 9(10): 28232-28255. doi: 10.3934/math.20241369
  • Many scholars have lately explored fractional-order boundary value issues with a variety of conditions, including classical, nonlocal, multipoint, periodic/anti-periodic, fractional-order, and integral boundary conditions. In this manuscript, the existence and uniqueness of solutions to sequential fractional differential inclusions via a novel set of nonlocal boundary conditions were investigated. The existence results were presented under a new class of nonlocal boundary conditions, Carathéodory functions, and Lipschitz mappings. Further, fixed-point techniques have been applied to study the existence of results under convex and non-convex multi-valued mappings. Ultimately, to support our findings, we analyzed an illustrative example.



    Lie groups and Lie algebras have become fundamental to pure mathematics and theoretical physics. The theory of Lie groups was first introduced and investigated by Sophus Lie through transformation groups[1]. The Killing vectors are used to construct new Lie groups. To obtain the Killing vectors, we solve a system of partial differential equations using an equation. After finding the Killing vectors, we can determine the Lie algebra by solving the Lie bracket, and then we can find the Lie group.

    A Lie group is a group that is also a differentiable manifold, meaning that it has a smooth structure compatible with the group operation. Lie groups are widely used in mathematics and physics to study geometric objects' symmetries, rigid bodies' rotations and translations, and the behavior of particles in quantum mechanics. One of their key properties is the Lie algebra, which is a vector space associated with the group that provides information about the group's infinitesimal behavior. They have since become a central object of study in modern mathematics and physics, with applications in fields ranging from algebraic geometry to string theory.

    Lie groups are extensively applied in various fields of mathematics and physics. Lie groups have wide-ranging applications in various fields of mathematics and physics, including differential geometry. Lie groups provide a natural setting for studying the geometry of curved spaces, making them a powerful tool in the field. For instance, the group of rotations in Euclidean space is a Lie group that plays a crucial role in understanding the geometry of rigid bodies. Through the study of Lie groups, mathematicians and physicists can develop a deeper understanding of the underlying structures and symmetries in various mathematical and physical systems. The rich mathematical theory of Lie groups continues to inspire new discoveries and insights, making it an active area of research in many branches of mathematics and physics.

    Moreover, Lie groups have broad and significant applications across various branches of mathematics and physics, such as group theory, physics, control theory, and quantum mechanics. In group theory, Lie groups play a crucial role in the study of properties of groups, including symmetry groups and permutation groups. In physics, Lie groups are commonly used to investigate symmetry and conservation laws, such as the Poincaré group, which is fundamental in studying special relativity. Control theory utilizes Lie groups to model complex dynamical systems like autonomous vehicles and robots. In the investigation of quantum mechanics, Lie groups play a crucial role in the theory of Lie algebras, which characterize the symmetry groups of physical systems. These symmetry groups are crucial in understanding the structure of these systems. Therefore, Lie groups have diverse and extensive applications in different fields of mathematics and physics, making them an essential tool for research and discovery.

    Differential geometry and mathematical physics use the term "Killing vector" to refer to a vector field on a smooth manifold that maintains the manifold's metric tensor. This vector field generates a symmetry transformation of the manifold's geometry. Specifically, Killing vectors are vector fields that preserve the geometry of a given space-time under an isometry (a certain type of transformation). In essence, a Killing vector generates a symmetry of the space-time while keeping its metric invariant. The Killing equation is the mathematical definition of a Killing vector as a vector field k that satisfies a specific condition

    μkν+νkμ=0, (2.1)

    where μ is the covariant derivative with respect to the metric of the space-time. The Killing equation is a differential equation that guarantees the vector field preserves the metric tensor, which in turn maintains the space-time geometry. To be classified as a killing vector, a vector field must fulfill various other conditions apart from the killing equation, such as being a smooth and continuous function on the space-time. Killing vectors are crucial in multiple fields of physics, such as general relativity and classical mechanics. They are employed to describe symmetries and conservation laws. The direction in which the Lie derivative of metric tensor is zero is called the Killing vector, and the Killing equation is

    Lξ(gab)=0 (2.2)

    and

    gab,cξc+gcbξc,a+gacξc,b=0, (2.3)

    where a comma (, ) represents a partial derivative. The Lie derivative, is a concept in differential geometry that has become a fundamental tool in modern physics and mathematics. The concept of Lie derivative of a tensor field along a vector field is a crucial notion in differential geometry. It measures the change of a tensor field along the flow defined by another vector field, taking into account the non-commutativity of the vector field. This directional derivative is widely used in physics and plays an indispensable role in understanding the behavior of physical systems, especially in the context of general relativity.

    The Lie derivative is also significant in defining the Lie algebra, which is the algebraic structure underlying the theory of Lie groups. Its far-reaching applications in various areas of physics and mathematics make it a crucial concept for examining the symmetries of physical systems and facilitating the development of mathematical models for various applications. Overall, the Lie derivative is a powerful mathematical operation that has become a cornerstone of modern physics and mathematics, providing insights into complex physical phenomena and helping to advance our understanding of the underlying mathematical structures.

    Section 1 is the introduction. Section 2 presents some preliminaries. Section 3 covers the Killing vectors of metric tensor. Section 4 covers Lie algebras of Killing vectors. Section 5 covers Lie group. Section 6 proves the formation of a regular surface. Section 7 covers the differential map for Lie group. Section 8 presents the differential of normal vector field on Lie group. Section 9 presents Gaussian and mean curvatures of the Lie group [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]. Finally, Section 10 contains the conclusions.

    This section focuses on the identification of the Killing vectors of the metric tensor of oblate spheroidal coordinates. Killing vectors are crucial concepts in differential geometry and physics that describe symmetries in space-time. They refer to vector fields that preserve the metric tensor of a manifold, a property that is significant in physics as it ensures the conservation of specific quantities such as energy and momentum along the flow of the Killing vector. The application of Killing vectors is seen in theories like general relativity and quantum field theory, where they help describe the symmetries of space-times such as black holes and the universe. Therefore, understanding the concept of Killing vectors is essential in helping physicists comprehend the symmetries of space-time and the fundamental laws of physics that govern it. Consider the well known metric tensor of oblate spheroidal coordinates[31],

    ds2=a2(sinh2(ξ)+sin2(η))d2ξ+a2(sinh2(ξ)+sin2(η))d2η+a2cosh2(ξ)cos2(η)d2ϕ,gab=(a2(sinh2(ξ)+sin2(η))000a2(sinh2(ξ)+sin2(η))000a2cosh2(ξ)cos2(η)),gab=(g11g12g13g21g22g23g31g32g33). (3.1)

    Let

    X=(X1(ξ,η,ϕ),X2(ξ,η,ϕ),X3(ξ,η,ϕ))

    be the Killing vectors of oblate spheroidal coordinates. From Eq (2.3), for g11

    g11,cXc+gc1Xc,1+g1cXc,1=0. (3.2)

    Since c = 1,2,3,

    g11,1X1+g11X1,1+g11X1,1+g11,2X2+g21X2,1+g12X2,1+g11,3X3+g31X3,1+g13X3,1=0, (3.3)
    (sinh2(ξ)+sin2(η))X1ξ+(sinh(ξ)cosh(ξ))X1+(sin(η)cos(η))X2=0. (3.4)

    For g12=g21,

    g12,cXc+gc2Xc,2+g1cXc,2=0, (3.5)
    g12,1X1+g12X1,1+g11X1,2+g12,2X2+g22X2,1+g12X2,2+g12,3X3+g32X3,1+g13X3,2=0, (3.6)
    X1η+X2ξ=0. (3.7)

    For g13=g31,

    g13,cXc+gc3Xc,1+g1cXc,3=0, (3.8)
    g13,1X1+g13X1,1+g11X1,3+g13,2X2+g23X2,1+g12X2,3+g13,3X3+g33X3,1+g13X3,3=0, (3.9)
    (sinh2(ξ)+sin2(η))X1ϕ+(cosh2(ξ)cos2(η))X3ξ=0. (3.10)

    For g22,

    g22,cXc+gc2Xc,2+g2cXc,2=0, (3.11)
    g22,1X1+g12X1,2+g21X1,2+g22,2X2+g22X2,2+g22X2,2+g22,3X3+g32X3,2+g23X3,2=0, (3.12)
    (sinh2(ξ)+sin2(η))X1η+sinh(2ξ)X1+(sinh2(ξ)+sin2(η))X2η+(sin(η)cos(η))X2=0. (3.13)

    For g23=g32,

    g23,cXc+gc3Xc,2+g2cXc,3=0, (3.14)
    g23,1X1+g13X1,2+g21X1,3+g23,2X2+g23,2X2+g22X2,3+g23,3X3,3+g33X3,2+g23X3,3=0, (3.15)
    (sinh2(ξ)+sin2(η))X2ϕ+(cosh2(ξ)cos2(η))X3η=0. (3.16)

    For g33,

    g33,cXc+gc3Xc,3+g3cXc,3=0, (3.17)
    g33,1X1+g13X1,3+g31X1,3+g33,2X2+g23X2+g32X2,3+g33,3X3,3+g33X3,3+g33X3,3=0, (3.18)
    X3ϕtan(η)X2+tanh(ξ)X1=0. (3.19)

    From Eq (3.4),

    X1ξ+(sinh(ξ)cosh(ξ)sinh2(ξ)+sin2(η))X1=0. (3.20)

    Then, we have

    X1=f(η,ϕ)sinh2(ξ)+sin2(η). (3.21)

    From Eq (3.13),

    (sinh2(ξ)+sin2(η))X2η+(sin(η)cos(η))X2=0, (3.22)
    X2η+(sinh(η)cos(η)sinh2(η)+sin2(η))X2=0. (3.23)

    Then, we have

    X2=g(η,ϕ)sinh2(ξ)+sin2(η). (3.24)

    From Eq (3.16),

    (sinh2(ξ)+sin2(η))ϕ(g(ξ,ϕ)sinh2(ξ)+sin2(η))+(cosh2(ξ)cos2(ξ))X3η=0, (3.25)
    sinh2(ξ)+sin2(η)g(ξ,ϕ)+(cosh2(ξ)cos2(ξ))X3η=0, (3.26)
    (cosh2(ξ)cos2(ξ))X3η=sinh2(ξ)+sin2(η)g(ξ,ϕ), (3.27)
    X3η=sinh2(ξ)+sin2(η)g(ξ,ϕ)(cosh2(ξ)cos2(ξ)). (3.28)

    Then, we have

    X3=2(sinh2(ξ)+sin2(η))323(cosh2(ξ)cos2(η)sin2(η)). (3.29)

    From Eq (3.32)

    η(f(η,ϕ)sinh2(ξ)+sin2(η))+ξ(g(η,ϕ)sinh2(ξ)+sin2(η))=0. (3.30)

    After solving the previous equations and comparing coefficients, we get

    f+f=0andgg=0.

    If

    f(0,0,0)=1andg(0,0,0)=1,

    then we obtain

    f=eη,   g=eξ. (3.31)

    By putting the value of f and g in (3.21) and (3.24), we obtain the Killing vectors

    X1=eηsinh2(ξ)+sin2(η), (3.32)
    X2=eξsinh2(ξ)+sin2(η), (3.33)
    X3=2(sinh2(ξ)+sin2(η))32eξ3(cosh2(ξ)cos2(η)sin2(η)), (3.34)

    X1,X2, and X3 are Killing vectors of oblate spheroidal coordinates.

    The main focus of this section to create the Lie algebra using oblate spheroidal coordinates' Killing vectors. To achieve this goal, the first step is to establish the Lie bracket of vector fields. The Lie bracket of Killing vectors is a critical concept in differential geometry that is crucial in the investigation of symmetries and conservation laws in physics. This concept provides information about the infinitesimal behavior of a Lie group and is utilized to determine the structure constants of the Lie algebra.

    The Lie algebra linked with the group is a valuable tool in understanding the symmetries of geometric objects and has widespread applications in physics and mathematics. Its applications range from studying the behavior of particles in quantum mechanics to analyzing the rotations and translations of rigid bodies, and examining the symmetries of space-times. The concept of Killing vectors is essential in differential geometry and mathematical physics. It refers to a vector field on a manifold that generates an isometry, preserving the metric tensor and describing a symmetry of the manifold that leaves distances between points unchanged.

    The Lie bracket of two Killing vectors is also a crucial object in the study of isometries of manifolds. It is defined as the commutator of their associated vector fields and has the property of being itself a Killing vector, which makes the space of Killing vectors on a given manifold a Lie algebra. This Lie algebra is a crucial tool in understanding the symmetries of a space. The Lie bracket of Killing vectors has applications in many areas of physics, including the study of conservation laws in general relativity, the construction of symmetries in quantum field theory, and the study of integrable systems in classical mechanics.

    Additionally, it is a fundamental tool in the study of geometric structures such as holonomy and curvature. Overall, the Lie bracket of Killing vectors is a powerful concept that plays a central role in the study of symmetries and geometry, with a wide range of applications in theoretical physics and mathematics.

    In the context of Lie algebras of Killing vectors, it is noted that the Lie bracket of two Killing vectors also results in a Killing vector. This is because Killing vectors uphold the metric tensor, and when a Killing vector is applied to the metric tensor, its Lie derivative is zero. Moreover, the Lie bracket operation on the space of Killing vectors satisfies the Jacobi identity, which is a fundamental property of Lie algebras. Specifically, for any three Killing vectors X,Y, and Z the Jacobi identity states that

    [X,[Y,Z]]+[Y,[Z,X]]+[Z,[X,Y]]=0, (4.1)

    where [X,Y] denotes the Lie bracket of X and Y[32]. The Lie algebra of Killing vectors plays an important role in the study of symmetries and conserved quantities in physics. For example, in general relativity, the presence of a Lie algebra of Killing vectors can be used to construct conserved quantities associated with the symmetries of spacetime. Similarly, in supersymmetric field theories, the Lie algebra of Killing vectors can be used to construct supersymmetric charges that are conserved quantities.

    Definition 4.1. [27] For two vector fields f(x) and g(x), the Lie bracket [f,g](x) is another vector field defined by

    [f,g](x)=gxf(x)fxg(x), (4.2)

    which is the Lie bracket of g along f.

    Now we find the Lie bracket of Killing vectors of oblate spheriodal coordinates.

    [X1,X2](x)=X2xX1(x)X1xX2(x), (4.3)
    X2x=eξ(2sinh2(ξ)+2sin2(η)sinh(2ξ)+sin(2η))2(sinh2(ξ)+sin2(η))32, (4.4)
    X2xX1=eξη(2sinh2(ξ)+2sin2(η)sinh(2ξ)+sin(2η))2(sinh2(ξ)+sin2(η))2, (4.5)
    X1x=eη(2sinh2(ξ)2sin2(η)sinh(2ξ)+sin(2η))2(sinh2(ξ)+sin2(η))32, (4.6)
    X1xX2=eξη(2sinh2(ξ)2sin2(η)sinh(2ξ)+sin(2η))2(sinh2(ξ)+sin2(η))2. (4.7)

    From Eqs (4.3), (4.5), and (4.7), we obtain

    [X1,X2]=2eξη(sinh2(ξ)+sin2(η)), (4.8)
    [X2,X3](x)=X3xX2(x)X2xX3(x), (4.9)
    X3x=eξ(sinh2(ξ)+sin2(η))12(3(sinh2(ξ)cosh2(η)sin2(η))2)1×[3sinh2(ξ)cosh2(η)sin2(η)(sinh(2ξ)sin(2η))+2(sinh2(ξ)+sin2(η))(sinh(2ξ)sinh(2η)sin(2η)], (4.10)
    X3xX2=e2ξ3(sinh2(ξ)cosh2(η)sin2(η))2×[3sinh2(ξ)cosh2(η)sin2(η)(sinh(2ξ)sin(2η))+2(sinh2(ξ)+sin2(η))(sinh(2ξ)sinh(2η)sin(2η))], (4.11)
    X2xX3=2e2ξ(sinh2(ξ)+sin2(η)sinh(ξ)cosh(ξ)+sin(η)cos(η))3(sinh2(ξ)cosh2(η)sin2(η)). (4.12)

    From Eqs (4.9), (4.11), and (4.12), we have

    [X2,X3]=e2ξ|[3sinh2(ξ)cosh2(η)sin2(η)|(sinh(2ξ)sin(2η)+2|(sinh2(ξ)+sin2(η)|)|)3sinh2(ξ)cosh2(η)sin2(η)||+2sinh(2ξ)sinh(2η)sin(2η)+2sinh2(ξ)+2sin2(η)sinh(2η)+sin(2η)3sinh2(ξ)cosh2(η)sin2(η)|], (4.13)
    [X1,X3](x)=X3xX1(x)X1xX3(x), (4.14)
    X3xX1=eξη|[3sinh2(ξ)cosh2(η)sin2(η)(sinh(2ξ)sin(2η))3sinh2(ξ)cosh2(ξ)sin2(η)||+2(sinh2(ξ)+sin2(η))(sinh(2ξ)sinh(2η)sin(2η))3sinh2(ξ)cosh2(ξ)sin2(η)|], (4.15)
    X1xX3=eξη(sinh2(ξ)sin2(η)sinh(ξ)cosh(ξ)+sin(η)cos(η))(sinh2(ξ)+sin2(η)). (4.16)

    From Eqs (4.14) and (4.16), we have

    [X1,X3]=eξη[3sinh2(ξ)cosh2(η)sin2(η)(sinh(2ξ)sin(2η))3sinh2(ξ)cosh2(ξ)sin2(η)+2(sinh2(ξ)+sin2(η))(sinh(2ξ)sinh(2ξ)sinh(2η)sin(2η))3sinh2(ξ)cosh2(ξ)sin2(η)]. (4.17)

    From Eqs (4.8), (4.13), and (4.17) we obtained a finite Lie algebras of Killing vectors.

    The objective of this section is to demonstrate that the Killing vectors being studied form a Lie group. A Lie group is a mathematical entity that merges the concepts of a group and a differentiable manifold. In other words, it is a combination of a group and a differentiable manifold, which allows it to be analyzed using calculus and characterized by smoothness or continuity. By analyzing the Killing vectors within the context of a Lie group, we can gain a better understanding of their properties and behavior.

    To provide more detail, a Lie group is a set (G) that comes equipped with a binary operation, usually denoted by multiplication, that satisfies the group axioms such as associativity, identity element, and inverse element. Additionally, it is also a smooth manifold, which means that it has a collection of charts that can describe it locally by coordinates. The smooth structure must be compatible with the group structure in the sense that the group multiplication operation must be a smooth function in the coordinates provided by the charts.

    Lie groups are essential mathematical objects that have many important applications in both mathematics and physics. One of their primary uses is in the study of symmetries of geometric objects and physical systems. For example, rotations, translations, and reflections in Euclidean space are all examples of Lie groups that arise from the study of continuous symmetries of geometric objects. Lie groups are abstract mathematical structures that have significant applications in both mathematics and physics. The general linear group and the special unitary group are examples of Lie groups that are commonly used in these fields. The study of Lie groups is a vast and complex subject with connections to various areas of mathematics and physics, including differential geometry, algebraic geometry, representation theory, and quantum field theory. Understanding the symmetries of physical systems and geometric objects is essential in both mathematics and physics, and Lie groups provide a fundamental tool to achieve this understanding. It is therefore crucial to study Lie groups for anyone interested in these fields.

    Theorem 5.1. The set of Killing vectors form a Lie group.

    Proof. Let

    G={X1,X2,X3}

    be set of Killing vectors, where X1,X2, and X3 are defined in Eqs (3.32)–(3.34), respectively. To prove G forms a Lie group, we need to define a group operation that satisfies certain properties. For this, we need to find a map that satisfies the conditions for a Lie group. The map that defines a Lie group for G is

    ϕ:R3R3byϕ(x,y,z)=(x1,y1,z1),

    where

    x1=ln(X1)+ln(X2),  x2=ln(X1)+ln(X2),  z1=z.

    We can show that G forms a Lie group under the map ϕ.

    (1) Closure

    The group operation is defined by the map ϕ, which takes a element (x,y,z)R3 and maps it to another element (x1,y1,z1)R3. Since expressions X1,X2, and X3 are continuous functions, the map ϕ is also continuous, which implies that the group operation is closed.

    (2) Associativity

    The group operation is defined by composition of maps which is associative.

    (3) Identity

    The identity element is the point (0,0,0) which corresponds to X1=1,X2=1, and X3=0.

    (4) Inverse

    The inverse of an element (x,y,z) is given by (x1,y1,z1). Since

    X1=ey1sinh2(x1)+sin2(y1)=eysinh2(x)+sin2(y)=(X1)1, (5.1)
    X2=ex1sinh2(x1)+sin2(y1)=exsinh2(x)+sin2(y)=(X2)1, (5.2)
    X3=2ex(sinh2(x1)+sin2(y1))323(sinh2(x1)cosh2(y1)sin2(y1)=X3. (5.3)

    (5) Smoothness

    The expressions X1,X2, and X3 are differentiable functions of their arguments and the map ϕ is also differentiable which implies that G is a differentiable manifold. Therefore, G forms a Lie group under the map ϕ.

    Theorem 5.2. Let X1,X2, and X3 be the Killing vectors. Then, these Killing vectors form a Lie group by using an exponential map.

    Proof. To define a Lie group, we need to define a Lie algebra generated by the killing vectors X1,X2, and X3 to a Lie group. One common way to do this is to use the exponential map which maps elements of the Lie algebra to elements of the Lie group. The exponential map is defined as exp: gG by

    exp(tX)=etX, (5.4)

    where t is a scalar parameter and X is a Killing vector. We can use this map to define the Lie group generated by the Killing vectors X1,X2, and X3.

    To do this, we need to express the Killing vectors X1,X2, and X3 in terms of a basis of the Lie algebra. One possible basis for this Lie algebra is given by

    [E1,E2,E3]=[X1,X2,X3]. (5.5)

    Using this basis, we can express any element of the Lie algebra as a linear combination of the basis elements

    (t1E1+t2E2+t3E3), (5.6)

    where t1,t2, and t3 are scalar parameters. To apply the exponential map to an element of the Lie algebra, we need to first find its matrix representation. This can be done by calculating the commutators of the basis elements

    [E1,E2]=0, (5.7)
    [E1,E3]=2E3, (5.8)
    [E2,E3]=2E3. (5.9)

    These commutation relations can be written as a matrix

    A=(000002020).

    The exponential map of an element of the Lie algebra can then be the found by exponentiating this matrix

    exp(t1E1+t2E2+t3E3)=e(t10+t20+t3A)=et3A, (5.10)

    where A is the matrix defined above. The resulting exponential map will define the elements of the Lie group generated by the Killing vectors X1,X2, and X3. Therefore, the Lie group generated by the Killing vectors X1,X2, and X3 can be defined using the exponential map as follows:

    g(x,y,z)=exp(xE1+yE2+zE3), (5.11)

    where x,y, and z are scalar parameters and E1,E2, and E3 are the basis elements of the Lie algebra defined above in terms of X1,X2, and X3.

    In this section, we prove that Lie group G form a regular surface.

    Definition 6.1. In differential geometry, a regular surface refers to a mathematical concept that describes a smooth two-dimensional object embedded in three-dimensional space. To be more specific, a regular surface is a collection of points in three-dimensional space, and every point in this surface has a neighborhood that can be described by a smooth map from a subset of the plane. This indicates that the surface is locally flat, and it has a well-defined tangent plane at each point. Spheres, cones, and tori are a few examples of regular surfaces. On the other hand, a surface that contains singularities, such as a cone with its tip or a sphere with a point removed, is not considered a regular surface.

    A surface patch X: UR3 is called regular if it is smooth and the vectors Xu and Xv are linearly independent at the point (u,v)U. Equivalently, X should be smooth and the vector product (Xu×Xv) should be non-zero at every point of U[33].

    Theorem 6.1. Let G be a Lie group defined in Theorem 5.1. This Lie group G forms a regular surface.

    Proof. Let

    G={X1,X2,X3}

    be a Lie group, where X1,X2, and X3 are defined in Eqs (3.32)–(3.34), respectively. In Figure 1, the representation of a regular surface is depicted by using these Killing vectors.

    G(ξ,η)=(eηsinh2(ξ)+sin2(η),eξsinh2(ξ)+sin2(η),2(sinh2(ξ)+sin2(η))32eξ3(cosh2(ξ)cos2(η)sin2(η))). (6.1)
    Figure 1.  Regular surface.

    Partially differentiate with respect to ξ,

    Gξ=(eη(sinh(ξ)cosh(ξ))(sinh2(ξ)+sin2(η))32,eξ(sinh2(ξ)+sin2(η)sinh(ξ)cosh(ξ))(sinh2(ξ)+sin2(η))32,2eξ(sinh2(ξ)+sin2(η))V(ξ,η)3(sinh2(ξ)cosh2(η)sin2(η))2), (6.2)
    V(ξ,η)=(sinh2(ξ)cosh2(η)sin2(η))(sinh2(ξ)+sin2(η))+3sinh3(ξ)cosh(ξ)cosh2(η)sin2(η)2(sinh2(ξ)+sin2(η))(sinh(ξ)cosh(ξ)cosh2(η)sin2(η)). (6.3)

    Take

    Gξ=(F1(ξ,η),F2(ξ,η),F3(ξ,η)). (6.4)

    Partially differentiate with respect to η,

    Gη=(eη(sinh2(ξ)sin2(η)+sin(η)cos(η))(sinh2(ξ)+sin2(η))32,eξ(sin(η)cos(η))(sinh2(ξ)+sin2(η))32,2eξ(sinh2(ξ)+sin2(η))12(3sinh2(ξ)cosh2(η)sin3(η)cos(η)+(sinh2(ξ)+sin2(η))sinh2(ξ)sinh(2ξ)sin(2η)3(sinh2(ξ)cosh2(η)sin2(η))2). (6.5)

    Take

    Gη=(G1(ξ,η),G2(ξ,η),G3(ξ,η)), (6.6)
    Gξ×Gη=(F2G3F3G2,F3G1F1G3,F1G2F2G1). (6.7)

    Since

    Gξ×Gη0,(ξ,η)R2 (6.8)

    and G is also smooth manifold. Therefore, Lie group G forms a regular surface.

    In the field of mathematics, a differential map refers to a function that connects two distinct manifolds, linking each point in the initial manifold to a linear transformation in the tangent spaces of the manifolds at that same point. Differential geometry is an area of mathematics that explores the study of geometric configurations on differential manifolds. The differential map is a key notion in differential geometry, representing a linear mapping between the tangent spaces of distinct differentiable manifolds.

    The study of differential geometry, which deals with the geometric characteristics of curved objects, relies heavily on the notion of a tangent space. At any given point on a manifold, the tangent space is defined as the collection of all possible velocities of curves that pass through that point. This allows us to comprehend the behavior of the manifold in a local context. Additionally, the differential map, also known as the derivative map or the pushforward map, is a potent tool for examining manifolds. It is a linear transformation that takes the tangent space at one point and maps it to the tangent space at another point. By contrasting the tangent spaces at various points on the manifold using the differential map, we can gain valuable insights into the geometry of the manifold. Furthermore, the characteristics of the differential map can reveal the local and global structure of the manifold, making it a crucial concept in the realm of differential geometry.

    The fundamental idea in differentiable manifolds is the concept of the differential map. This map characterizes the linearization of a differentiable function f: MN between two manifolds, M and N, at a point p. Essentially, the differential map is a linear transformation that maps the tangent space TpM of M at p to the tangent space Tf(p)N of N at f(p). It describes the relationship between the tangent spaces of M and N through the map f, enabling us to relate geometric concepts between manifolds that are connected through differentiable maps. Geometrically, the differential map maps a tangent vector vTpM to a tangent vector dfp(v)Tf(p)N, indicating the image of the tangent vector v under the influence of f at p. This makes the differential map a powerful tool for comprehending the local geometry of maps between manifolds and comprehending the structure and behavior of differentiable maps.

    The differential map is an essential tool in differential geometry, as it allows us to compare geometric structures on differentiable manifolds. By using the differential map, we can define the concept of a smooth map between two manifolds, M and N. A smooth map is a differentiable function f: MN that preserves the tangent structure of the manifold. In other words, f is a smooth map if it maps tangent vectors in M to tangent vectors in N in a smooth and consistent manner. The differential map is employed to define smoothness rigorously by assessing how well the map preserves the tangent spaces of the manifolds. Therefore, the differential map is a crucial concept in differential geometry that plays a significant role in studying the local and global properties of differentiable manifolds.

    To summarize, the differential map is a crucial concept in differential geometry that serves as a key link between the geometry of differentiable manifolds. By mapping tangent vectors from one manifold to another, the differential map enables us to compare the local geometry of the manifolds and explore their interrelationships. Additionally, the differential map is employed to define essential concepts in differential geometry, such as smooth maps and map differentials. Ultimately, the differential map is a potent tool that aids us in comprehending the structure of differentiable manifolds and their connections through linear maps between their tangent spaces.

    Definition 7.1. [33] Let f: SRn be a differentiable map and pS, vTp(S). Then, there exists a curve α: (ϵ,ϵ)S such that α(0)=p, α(0)=v and map dfp: Tp(S1)Tf(p)(S2)

    dfp(v)=ddt(fα)(t),t=0, (7.1)
    dfp(v)=ddt(f(α(t)),t=0. (7.2)

    Theorem 7.1. If f: SRn is a differentiable map. Then, there exists a differential of f for the Lie group G defined above in Theorem 5.1 is dfp(v).

    Proof. Let

    α(t)=(a1(t),a2(t),a3(t))

    be the curve passing through a point p on the surface G. Take differentiate with respect to t,

    α(t)=(a1(t),a2(t),a3(t)). (7.3)

    Define a map α: (ϵ,ϵ)S such that α(0)=p, α(0)=v

    v=α(0)=(a1(0),a2(0),a3(0)), (7.4)

    so,

    dfp(v)=ddt(fα)(t),t=0, (7.5)
    dfp(v)=ddt(f(α(t))) (7.6)
    dfp(v)=ddt(f(a1(t),a2(t),a3(t)), (7.7)
    dfp(v)=ddt(ea1(t)sinh2(a1(t)+sin2(a1(t),ea2(t)sinh2(a2(t))+sin2(a2(t)),2(sinh2(a3(t))+sin2(a3(t)))32ea3(t)3(sinh2(a3(t)cosh2(a3(t))sin2(a3(t))), (7.8)
    dfp(v)=(a1(t)ea1(t)(2sinh2(a1(t))2sin2(a1(t))sinh(2a1(t))+sin(2a1(t)))2(sinh2(a1(t))+sin2(a1(t)))32a2(t)ea2(t)(2sinh2(a2(t))+2sin2(a2(t))sinh(2a2(t))+sin(2a2(t)))2(sinh2(a2(t))+sin2(a2(t)))323a3(t)ea3(t)(sinh2(a3(t))cosh2(a3(t))sin2(a3(t)))U(t)3(sinh2(a3(t))cosh2(a3(t))sin2(a3(t)))2), (7.9)

    where

    U(t)=sinh2(a3(t))+sin2(a3(t))(sinh(2a3(t))sin(2a3(t)))+2(sinh2(a3(t))+sin2(a3(t)))32(sinh(2a3(t))sinh(2a3(t))sin(2a3(t))), (7.10)

    which is required differential map for Lie group G.

    The differential of a unit normal vector field is an important concept in differential geometry that helps us to understand how the normal vector to a surface varies as we traverse the surface. More specifically, if N is a unit normal vector field on a smooth surface S in three-dimensional space, the differential of N at a point p on S is a vector field that describes the infinitesimal changes in the normal vector as we move around on the surface.

    This concept is vital in differential geometry since it gives us valuable information about the geometry and curvature of the surface. The differential of a normal vector field is employed in various applications, including the study of geometric flows, surface dynamics, and physical laws that involve the behavior of surfaces in space. Therefore, the differential of a normal vector field is a significant concept in differential geometry, offering crucial insights into the local and global properties of surfaces in three-dimensional space.

    Definition 8.1. [33] Let a map N: SS2 which is a Gauss map. The differential of N is defined by

    dNp:Tp(S)TN(p)(S2)=Tp(S),

    where N is a normal vector field on surface S

    dNp(v)=ddt(Nα)(t),t=0, (8.1)

    where α is a curve passing through the point on the surface.

    Theorem 8.1. Let N be a unitary normal vector field. Then, the differential of normal vector field of the surface which is a Lie group is

    dNp(v)=vR,t=0. (8.2)

    Proof. Since the unit normal vector field is

    N=(F2G3F3G2,F3G1F1G3,F1G2F2G1)(F2G3F3G2)2+(F3G1F1G3)2+(F1G3F2G1)2. (8.3)

    Let

    α(t)=(a1(t),a2(t),a3(t))

    be the curve passing through a point p on the surface G. Take differentiate with respect to t

    α(t)=(a1(t),a2(t),a3(t)). (8.4)

    Define a map α: (ϵ,ϵ)S, such that α(0)=p, α(0)=v,

    v=α(0)=(a1(0),a2(0),a3(0)), (8.5)
    dNp(v)=ddt(Nα)(t),t=0, (8.6)
    dNp(v)=ddt(N(α(t)), (8.7)
    dNp(v)=ddt(N(a1(t),a2(t),a3(t))), (8.8)
    dNp(v)=ddt((F2(a1(t))G3(a1(t))G2(a1(t))F3(a1(t)))R)ddt((G1(a2(t))F3(a2(t))F1(a2(t))G3(a2(t)))R)ddt((F1(a3(t))G2(a3(t))F2(a3(t))G1(a3(t)))R), (8.9)

    where

    R=(F2G3F3G2)2+(F3G1F1G3)2+(F1G3F2G1)2,
    dNp(v)=1R[F2(a1(t))G3(a1(t))+F2(a1(t))G3(a1(t))G2(a1(t))F3(a1(t))G2(a1(t))F3(a1(t))G1(a2(t))F3(a2(t))+G1(a2(t))F3(a2(t))F1(a2(t))G3(a2(t))F(a2(t))G3(a2(t))F1(a3(t))G2(a3(t))+F1(a3(t))G2(a3(t))F2(a3(t))G1(a3(t))F2(a3(t))G1(a3(t))], (8.10)
    dNp(v)=vR,t=0, (8.11)

    which is required.

    In this section, the main objective is to determine the Kg and H of a surface that is classified as a Lie group. The methodology used for this task involves calculating the first fundamental form (FFF) and second fundamental form (SFF), which allow us to obtain the coefficients needed to determine the curvatures. It is important to note that this particular approach is specific to surfaces that fall under the category of Lie groups, and cannot be applied to other types of surfaces. By employing this method, we can gain a comprehensive understanding of the surface's geometry and properties. Overall, this process of determining curvatures serves as a valuable tool for further analysis and investigation of surfaces.

    The FFF is a crucial mathematical tool that is extensively used to analyze surfaces in three-dimensional space. It enables the measurement of the intrinsic geometry of a surface, independent of any particular embedding in three-dimensional space. To define the FFF, the dot product of the tangent vectors to a surface at a given point is computed. This measures the inner product of the surface's partial derivatives with respect to its two parameters. The dot product provides essential information about the angles and lengths of curves on the surface, which are fundamental in comprehending the surface's geometry. Researchers can gain a better understanding of surfaces and their properties by utilizing the FFF.

    The FFF is an essential concept in the field of differential geometry, which plays a vital role in the calculation of different geometric properties of surfaces. Mathematicians use this concept to determine surface curvature, length of curves on the surface, and other related properties. It is a mathematical tool represented by the symbol I, and it is defined as the dot product of tangent vectors to a surface at a particular point. The application of the FFF is not limited to the study of surfaces but also encompasses the investigation of minimal surfaces that minimize their surface area subject to certain constraints. This mathematical concept is crucial in understanding the local geometry of a surface in three-dimensional Euclidean space. In conclusion, the FFF is a fundamental tool in differential geometry that plays a critical role in the study of surfaces and is represented by an equation[33]

    I=Edu2+2Fdudv+Gdv2, (9.1)

    where u and v are the parameters used to parameterize the surfac; and E,F, and G are functions of u and v that describe the local geometry of the surface. The coefficients of FFF are a vital aspect of surface analysis. In particular, the terms E,F, and G correspond to the lengths of the edges of the tangent plane to the surface and play a critical role in understanding the surface's geometry. Specifically, E and G represent the squares of the lengths of the two tangent vectors, u and v, respectively. On the other hand, F represents their dot product. By utilizing these coefficients, mathematicians can better understand the local geometry of a surface, and determine its curvature and other geometric properties. Therefore, understanding the meaning and significance of E,F, and G is fundamental in the study of surfaces and their properties. In other words, the first fundamental form describes how the surface bends and curves locally, and it is an essential tool for studying properties of surfaces, such as their curvature and area. The coefficients of FFF are

    E=Gu.Gu=∥Gu2, (9.2)
    E=F21+F22+F23, (9.3)
    F=Gu.Gv=F1G1+F2G2+F3G3, (9.4)
    G=Gv.Gv=G21+G22+G23. (9.5)

    FFF is

    I=Edu2+2Fdudv+Gdv2, (9.6)
    I=(F21+F22+F33)du2+2(F1G1+F2G2+F3G3)dudv+(G21+G22+G23)dv2. (9.7)

    The L,M, and N coefficients play a crucial role in calculating different geometric quantities related to a surface. For instance, they can be used to determine Kg and H of the surface. The mean curvature, in particular, can be obtained by computing the trace of the SFF, which is equivalent to adding up the L,M, and N coefficients. It is important to note that these coefficients are independent of the coordinate system used and can provide valuable insights into the shape and behavior of a surface.

    Understanding the SFF is important in many areas of mathematics and physics, such as the study of minimal surfaces, the theory of relativity, and the mechanics of surfaces under stress. The coefficients of SFF are

    Guu=(F1(u,v),F2(u,v),F3(u,v)),  (9.8)
    Guv=(H1(u,v),H2(u,v),H3(u,v)), (9.9)
    Gvv=(G1(u,v),G2(u,v),G3(u,v)). (9.10)

    The unitary normal vector is

    P=(F2G3F3G2,F3G1F1G3,F1G2F2G1)(F2G3F3G2)2+(F3G1F1G3)2+(F1G3F2G1)2, (9.11)

    so

    L=Guu.P, (9.12)
    L=1R[F1(F2G3F3G2)+F2(F3G1F1G3)+F3(F1G2F2G1)], (9.13)

    where (') represents the derivative.

    M=Guv.P, (9.14)
    M=1R[H1(F2G3G2F3)+H2(G1F3F1G3)+H3(F1G2G1F2)], (9.15)
    N=Gvv.P, (9.16)
    N=1R[G1(F2G3G2F3)+G2(G1F3F1G3)+G3(F1G2G1F2)]. (9.17)

    The SFF is

    II=Ldu2+2Mdudv+Ndv2, (9.18)
    II=1R[F1(F2G3G2F3)+F2(G1F3F1G3)+F3(F1G2G1F2)]du2+2R[H1(F2G3G2F3)+H2(G1F3F1G3)+H3(F1G2G1F2)]dudv+1R[G1(F2G3G2F3)+G2(G1F3F1G3)+G3(F1G2G1F2)]dv2. (9.19)

    The Kg and H of a surface are geometric quantities that can be computed using the FFF and SFF coefficients of the surface. The FFF coefficients of a surface are the coefficients of the metric tensor, which determine the lengths of tangent vectors on the surface. The SFF coefficients describe how the surface curves in the ambient space. Specifically, they are related to the shape of the surface in the normal direction. The measurement of how much a surface curves in two dimensions is determined by its Kg. This curvature is calculated as the product of the surface's principal curvatures and can be computed using the coefficients of the FFF and SFF. To be precise, the Kg can be derived from the FFF coefficients, denoted as E,F, and G and the SFF coefficients (L,M, and N)[33]

    Kg=(LNM2)/(EGF2). (9.20)

    The mean curvature of a surface is a measure of how much the surface curves in one dimension. It is defined as the average of the principal curvatures of the surface, which can also be computed using the FFF and SFF coefficients. Specifically, the mean curvature is given by

    H=(1/2)((EN+GL2FM)/(EGF2)), (9.21)

    so, the Kg and H of a surface can be computed using the FFF and SFF coefficients.

    LNM2=1R2[(F1(F2G3F3G2)+F2(G1F3F1G3)+F3(F1G2G1F2))+G1(F2G3F3G2)+G2(F3G1F1G3)+G3(F1G2F2G1)(H1(F2G3F3G2)+H2(F3G1F1G3)+H3(F1G2F2G1))2]. (9.22)

    Take

    LNM2=Q(F1,F2,F3,G1,G2,G3,H1,H2,H3,F1,F2,F3,G1,G2,G3), (9.23)
    GEF2=(G21+G22+G33)(F21+F22+F23)(F1G1+F2G2+F3G3)2. (9.24)

    Take

    GEF2=W(F1,F2,F3,G1,G2,G3), (9.25)

    so,

    Kg=Q(F1,F2,F3,G1,G2,G3,H1,H2,H3,F1,F2,F3,G1,G2,G3)W(F1,F2,F3,G1,G2,G3). (9.26)

    The mean curvature is

    H=GL+NE2MF2(GEF2), (9.27)
    GL+NE2MF=1R((G21+G22+G23)[F1(F2G3F3G2)+F2(G1F3F1G3)+F3(F1G2F2G1)]+(F21+F22+F23)[G1(F2G3F3G2)+G2(F3G1F1G3)+G3(F1G2F2G1)]2(F1G1+F2G2+F3G3)×[H1(F2G3F3G2)+H2(F3G1F1G3)+H3(F1G2F2G1)]). (9.28)

    Take

    GL+NE2MF=Z(F1,F2,F3,G1,G2,G3,H1,H2,H3,F1,F2,F3,G1,G2,G3), (9.29)

    so,

    H=Z(F1,F2,F3,G1,G2,G3,H1,H2,H3,F1,F2,F3,G1,G2,G3)2W(F1,F2,F3,G1,G2,G3). (9.30)

    In this paper, we constructed a new Lie group with the help of Killing vectors. First, we find the Killing vectors and then constructed the Lie algebra by using the Lie bracket on Killing vectors. After we constructed the Lie algebra, we proved that these Killing vectors form a Lie group by using an exponential map. Then, we discussed its geometric properties. We defined the differential map and differential of normal vector on the Lie group and also calculated its Gaussian and mean curvatures.

    The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.

    The authors thank to Dr. Ibtesam Eid Almasoudi Alshammari, Vice dean of E- Learning and Digital Transformation, Associate Professor, Faculty of Science, University of Hafr Al Batin for funding this research work.

    The authors declare no conflicts of interest.



    [1] B. Ahmad, S. K. Ntouyas, Existence results for Caputo-type sequential fractional differential inclusions with nonlocal integral boundary conditions, J. Appl. Math. Comput., 50 (2016), 157–174. https://doi.org/10.1007/s12190-014-0864-4 doi: 10.1007/s12190-014-0864-4
    [2] M. Kisielewicz, Stochastic differential inclusions and applications, In: Springer optimization and its applications, New York: Springer, 2013. https://doi.org/10.1007/978-1-4614-6756-4
    [3] H. A. Hammad, R. A. Rashwan, A. Nafea, M. E. Samei, M. De la Sen, Stability and existence of solutions for a tripled problem of fractional hybrid delay differential euations, Symmetry, 14 (2022), 2579. https://doi.org/10.3390/sym14122579 doi: 10.3390/sym14122579
    [4] H. A. Hammad, R. A. Rashwan, A. Nafea, M. E. Samei, S. Noeiaghdam, Stability analysis for a tripled system of fractional pantograph differential equations with nonlocal conditions, J. Vib. Control, 30 (2024), 632–647. https://doi.org/10.1177/10775463221149232 doi: 10.1177/10775463221149232
    [5] M. F. Danca, Synchronization of piecewise continuous systems of fractional order, Nonlinear Dyn., 78 (2014), 2065–2084. https://doi.org/10.1007/s11071-014-1577-9 doi: 10.1007/s11071-014-1577-9
    [6] Y. Cheng, R. P. Agarwal, D. O. Regan, Existence and controllability for nonlinear fractional differential inclusions with nonlocal boundary conditions and time-varying delay, FCAA, 21 (2018), 960–980. https://doi.org/10.1515/fca-2018-0053 doi: 10.1515/fca-2018-0053
    [7] S. K. Ntouyas, S. Etemad, J. Tariboon, Existence results for multi-term fractional differential inclusions, Adv. Diff. Equ., 2015 (2015), 140. https://doi.org/10.1186/s13662-015-0481-z doi: 10.1186/s13662-015-0481-z
    [8] A. Losta, Z. Opial, Application of the Kakutani-Ky-Fan theorem in the theory of ordinary differential equations or noncompact acyclic-valued map, Bull. Acad. Pol. Sci., Ser. Sci. Math. Astron. Phys., 13 (1965), 781–786.
    [9] S. Chakraverty, R. M. Jena, S. K. Jena, Computational fractional dynamical systems: Fractional differential equations and applications, John Wiley & Sons, 2022. https://doi.org/10.1002/9781119697060
    [10] B. Ahmad, V. Otero-Espinar, Existence of solutions for fractional differential inclusions with antiperiodic boundary conditions, Bound. Value Probl., 2009 (2009), 625347. https://doi.org/10.1155/2009/625347 doi: 10.1155/2009/625347
    [11] H. A. Hammad, M. Zayed, Solving systems of coupled nonlinear Atangana-Baleanu-type fractional differential equations, Bound. Value Probl., 2022 (2022), 101. https://doi.org/10.1186/s13661-022-01684-0 doi: 10.1186/s13661-022-01684-0
    [12] H. A. Hammad, H. Aydi, M. Zayed, On the qualitative evaluation of the variable-order coupled boundary value problems with a fractional delay, J. Inequal. Appl., 2023 (2023), 105. https://doi.org/10.1186/s13660-023-03018-9 doi: 10.1186/s13660-023-03018-9
    [13] Humaira, H. A. Hammad, M. Sarwar, M. De la Sen, Existence theorem for a unique solution to a coupled system of impulsive fractional differential equations in complex-valued fuzzy metric spaces, Adv. Differ. Equ., 2021 (2021), 242. https://doi.org/10.1186/s13662-021-03401-0 doi: 10.1186/s13662-021-03401-0
    [14] M. Subramanian, M. Manigandan, C. Tunc, T. N. Gopal, J. Alzabut, On the system of nonlinear coupled differential equations and inclusions involving Caputo-type sequential derivatives of fractional order, J. Taibah Uni. Sci., 16 (2022), 1–23. https://doi.org/10.1080/16583655.2021.2010984 doi: 10.1080/16583655.2021.2010984
    [15] M. Manigandan, S. Muthaiah, T. Nandhagopal, R. Vadivel, B. Unyong, N. Gunasekaran, Existence results for a coupled system of nonlinear differential equations and inclusions involving sequential derivatives of fractional order, AIMS Math., 7 (2022), 723–755. https://doi.org/10.3934/math.2022045 doi: 10.3934/math.2022045
    [16] A. Perelson, Modeling the interaction of the immune system with HIV, In: Mathematical and statistical approaches to AIDS epidemiology, Berlin: Springer, 1989,350–370. https://doi.org/10.1007/978-3-642-93454-4_17
    [17] A. S. Perelson, D. E. Kirschner, R. De Boer, Dynamics of HIV infection of CD4+ T cells, Math. Biosci., 114 (1993), 81–125. https://doi.org/10.1016/0025-5564(93)90043-a doi: 10.1016/0025-5564(93)90043-a
    [18] D. Baleanu, O. G. Mustafa, R. P. Agarwal, On Lp-solutions for a class of sequential fractional differential equations, Appl. Math. Comput., 218 (2011), 2074–2081. https://doi.org/10.1016/j.amc.2011.07.024 doi: 10.1016/j.amc.2011.07.024
    [19] A. A. Kilbas, H. M. Srivastava, J. J. Trujillo, Theory and applications of fractional differential equations, Elsevier, 2006.
    [20] K. S. Miller, B. Ross, An introduction to the fractional calculus and fractional differential equations, Wiley, 1993.
    [21] Z. Wei, W. Dong, Periodic boundary value problems for Riemann-Liouville sequential fractional differential equations, Electron. J. Qual. Theory Differ. Equ., 87 (2011), 1–13. https://doi.org/10.14232/ejqtde.2011.1.87 doi: 10.14232/ejqtde.2011.1.87
    [22] H. A. Hammad, H. Aydi, H. Isik, M. De la Sen, Existence and stability results for a coupled system of impulsive fractional differential equations with Hadamard fractional derivatives, AIMS Math., 8 (2023), 6913–6941. https://doi.org/10.3934/math.2023350 doi: 10.3934/math.2023350
    [23] H. A. Hammad, M. De la Sen, Stability and controllability study for mixed integral fractional delay dynamic systems endowed with impulsive effects on time scales, Fractal Fract., 7 (2023), 92. https://doi.org/10.3390/fractalfract7010092 doi: 10.3390/fractalfract7010092
    [24] X. Li, D. Chen, On solvability of some p-Laplacian boundary value problems with Caputo fractional derivative, AIMS Math., 6 (2021), 13622–13633. https://doi.org/10.3934/math.2021792 doi: 10.3934/math.2021792
    [25] H. A. Hammad, P. Agarwal, S. Momani, F. Alsharari, Solving a fractional-order differential equation using rational symmetric contraction mappings, Fractal Fract., 5 (2021), 159. https://doi.org/10.3390/fractalfract5040159 doi: 10.3390/fractalfract5040159
    [26] Z. Wei, Q. Li, J. Che, Initial value problems for fractional differential equations involving Riemann-Liouville sequential fractional derivative, J. Math. Anal. Appl., 367 (2010), 260–272. https://doi.org/10.1016/j.jmaa.2010.01.023 doi: 10.1016/j.jmaa.2010.01.023
    [27] M. Klimek, Sequential fractional differential equations with Hadamard derivative, Commun. Nonlinear Sci., 16 (2011), 4689–4697. https://doi.org/10.1016/j.cnsns.2011.01.018 doi: 10.1016/j.cnsns.2011.01.018
    [28] C. Bai, Impulsive periodic boundary value problems for fractional differential equation involving Riemann-Liouville sequential fractional derivative, J. Math. Anal. Appl., 384 (2011), 211–231. https://doi.org/10.1016/j.jmaa.2011.05.082 doi: 10.1016/j.jmaa.2011.05.082
    [29] Y. K. Chang, J. J. Nieto, Some new existence results for fractional differential inclusions with boundary conditions, Math. Comput. Model., 49 (2009), 605–609. https://doi.org/10.1016/j.mcm.2008.03.014 doi: 10.1016/j.mcm.2008.03.014
    [30] S. Hu, N. S. Papageorgiou, Handbook of multivalued analysis (theory), Dordrecht: Kluwer Academic Publishers, 1997.
    [31] K. Deimling, Multivalued differential equations, New York: De Gruyter, 1992. https://doi.org/10.1515/9783110874228
    [32] I. Podlubny, Fractional differential equations: An introduction to fractional derivatives, fractional differential equations, methods of their solution, and some of their applications, Elsevier, 1998.
    [33] A. Granas, J. Dugundji, Fixed point theory, New York: Springer, 2003.
    [34] H. Covitz, S. B. Nadler, Multi-valued contraction mappings in generalized metric spaces, Israel J. Math., 8 (1970), 5–11. https://doi.org/10.1007/BF02771543 doi: 10.1007/BF02771543
    [35] C. Castaing, M. Valadier, Convex analysis and measurable, multifunctions, In: Lecture Notes in Mathematics, Berlin: Springer, 1977. https://doi.org/10.1007/BFb0087685
  • Reader Comments
  • © 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1053) PDF downloads(69) Cited by(4)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog