dc.description.abstract |
Evolution equation arises in di erent elds of science and engineering, such as
the propagation of waves, generic repression, laser optics, whaling control, control
theory, climate models, reaction di usion equations, coupled oscillators, population
ecology, viscosity materials, food webs, structured population models, enzyme kinetics,
neural networks, modern physics and biology, can be modelled as abstract
functional and neutral functional di erential equations in Hilbert spaces or more generally,
in Banach spaces, where the space variable is suppressed and the equation
looks like an ordinary di erential equation in time variable in an in nite dimensional
space. If we are concerned with the invariant properties of certain classes
of problems, then the best way to study such problems is to consider the abstract
formulations of such problems. By an abstract formulation, we indicate a functional
analytic representation of the problems. The evolution equations are the abstract
formulation of the prototype of many problems. In in nite dimensional space, ordinary
di erential equations can be regarded as evolution equations.
Functional di erential equations may give a mathematical model for a physical
system in which the rate of progress of the system may rely on its past history, that
is, the future state of the system depends on the present as well as a part of its past
history. The functional di erential equation with nonlocal conditions is considered
to be a valuable tool in the modeling of many phenomena in various areas of sciences,
engineering and economics due to their practical application to many physical
v
problems(population dynamics, pollution process in river and sea caused by sewage
and many others) and hence they have picked up impressive consideration.
Mathematical formulation of nonlocal problems also arises naturally in various
engineering models, such as, heat conduction, semiconductor modelling, nonlocal
reactive transport in underground waterows in porous media and biotechnology.
The quantity of physical phenomena displayed by partial di erential equations with
nonlocal conditions which haves abstract formulation as a functional di erential
equation is continually expanding. These facts emphasize the importance of examining
the nonlocal problems being considered as a model of a physical system for
the existence, uniqueness and regularity results.
The fractional calculus, which provides the di erentiation and integration of arbitrary
order, has been a productive eld of research in many area of sciences and
engineering such as, in viscoelasticity and damping, electromagnetism, di usion
and wave propagation, chaos and fractals, biology, electronics, signal processing,
robotics, tra c systems, genetic algorithms, percolation, chemistry, irreversibility,
physics, control systems as well as economy, nance and many others. In fact,
fractional di erential equations are considered as an alternative model to nonlinear
di erential equations. The most important advantage of utilizing them is their nonlocal
property. It is well realized that the fractional order di erential operator is
non-local but the integer order di erential operator is a local operator.
Di erential inclusions are used for describing systems with hysteresis. All the
problems considered for di erential equation, i.e. existence of mild solutions, continuation
of solutions, dependence on initial conditions and parameters, are present
in the theory of di erential inclusion. Since a di erential inclusion usually has many
solutions starting at a given point, new issues may appear, for example, study of
topological properties of the set of solutions, optimal control theory, selection of
vi
solutions with given properties evaluation of the reachability sets and so on. However,
the di erential inclusions are not only models for many dynamical processes
but also, they present a powerful tool for various branches of mathematical analysis.
Di erential inclusion is used to derive su cient conditions of optimality, play an
essential role in the theory of control under conditions of uncertainty and in di erential
game theory.
A stochastic di erential equation of fractional order is a fractional di erential
equation containing random term/s (noise term). The investigation of stochastic
fractional di erential equations is an energizing topic which unites procedures from
probability hypothesis, functional analysis, and the hypothesis of fractional partial
di erential equations. In the recent decades, the investigation of stochastic partial
di erential equations has turned into a standout amongst the most quickly extending
ranges in probability theory. In addition to applications to various issues in
mathematical physics and life sciences, enthusiasm for such studies is motivated by
a desire to understand and control the behavior of complex systems that show up in
numerous zones of natural and social sciences. Thus, there has been a great deal of
interest in optimal control systems described by stochastic and partial di erential
equations. These optimal control problems lead to stochastic and partial di erential
inclusions.
Several evolutionary operations from areas as diverse as population dynamics,
orbital transfer of satellites, sampled-data systems and engineering are portrayed
by the fact that they undergo abrupt changes. The total length of such changes is
negligible in comparison with the entire duration of the process and thus the abrupt
changes can be well-approximated in terms of instantaneous changes of the state i.e.
impulses. Such process can be modeled by impulsive di erential equations allowing
for discontinuous in the evolution of the state. Impulsive di erential equations are
vii
usually de ned by a pair of equations to be satis ed during the continuous portions
of evolution and a di erence equation de ning the discrete impulsive action. If some
spatio-temporal relation satis es, then impulses occur. Particular impulsive di erential
equations simulating the work of concrete systems have been investigated by
many authors. Also, some biological events can be appropriately described by impulsive
di erential equations [31], [214].
The main object of thesis is to study of numerous type of di erential equations
of integer and fractional order, functional integro-di erential equations of the
rst order, neutral fractional di erential equations with nite delay or in nite delay,
nonlocal stochastic fractional integro-di erential inclusion with impulses, and
investigate the existence of mild solution such considered systems utilizing di erent
techniques.
Chapter 1 provides some introductory matter and literature survey of the problems
considered in the subsequent chapters giving an inspiration to the study provided
in the thesis. Some elementary de nitions and results, which are used in
subsequent chapters, are considered in Chapter 2.
Chapter 3 contains two problems and deals with the semi-linear di erential
equations in a Banach space involving non-instantaneous impulses. By using xed
point theorems for convex-power condensing, the su cient conditions for the existence
of the mild solution is obtained via the techniques of the measure of noncompactness
and analytic semigroup without assuming Lipschitz continuity of nonlinear
function. For the second problem, we also obtain the su cient condition to prove
the existence of positive mild solution. For illustrating the obtained theory, we consider
the examples.
With the help of Hausdor measure of noncompactness and fractional operators,
viii
Chapter 4 studies an impulsive neutral integro-di erential equation with in nite delay
in Banach space. Utilizing Darbo-Sadovskii xed point theorem, the existence of
the mild solution is established without assuming Lipschitz continuity of nonlinear
function. In order to prove required result, we need only equicontinuity of resolvent
operator. That's why, we consider the analytic resolvent operator. An application
is also provided at the end of the chapter.
Chapter 5 investigates the existence of the mild solution for the fractional
di erential equation with state-dependent delay in Banach space. The su cient
conditions, providing existence of a mild solution are given by means of resolvent
operator and some xed point theorems. Finally, some examples are considered to
illustrate the discussed theory.
An impulsive di erential equation of fractional order with a deviated argument
is under consideration in Chapter 6. We study the approximate solution of considered
impulsive di erential equation with deviating argument via the technique
of Faedo-Galerkin approximations and analytic semigroup method. The Faedo-
Galerkin method will result in a nite-dimensional approximation of the evolution
equation providing more regular solutions under weaker assumptions. First of all
we provide preliminaries and assumptions required in this chapter. We study an
associated integral equation and then consider a sequence of approximate integral
equations obtained by applying the projection operator on the associated integral
equation. Utilizing Banach xed point theorem and semigroup theory, we study the
existence and uniqueness of each approximate integral equation. Then, we show the
convergence of approximate integral equation to limiting function that satis es the
associated integral equation. Next, we consider the Faedo-Galerkin approximation
and demonstrate some convergence results with the help of it. Finally, we give an
ix
example showing the e ectiveness of obtained abstract results.
Following the idea of Chapter 6, Chapter 7 studies the Faedo-Galerkin approximations
of solutions to a nonlocal neutral di erential equations of fractional order in
a separable Hilbert space. We establish the existence and uniqueness of the solution
to every approximate integral equation obtained by applying the projection operator
on associate integral equation via xed point technique and semigroup theory.
To illustrate the discussed abstract theory, we give an example at the end of the
Chapter.
An impulsive neutral stochastic integro-di erential equation of fractional order
with in nite delay is studied in the Chapter 8. This chapter discusses two existence
results. The su cient condition to prove the rst existence result providing existence
and uniqueness of the mild solution is derived via the technique of Banach
xed point theorem and resolvent operator method. The second result is obtained
by utilizing Krasnoselskii-Schaefer xed point theorem with compact resolvent operator.
An example is also considered illustrating the given theory.
The closing Chapter 9 is devoted to study a nonlocal stochastic fractional
integro-di erential inclusion with impulsive conditions in a real separable Hilbert
space. The existence of a mild solution is established by utilizing a xed point theorem
for multi-valued operators due to Dhage and resolvent operator with fractional
power operator theory. An example is also provided to show the applicability of
obtained results.
The relevant references are appended at the end. |
en_US |