The perturbation theory plays an important role in studying structure formation in cosmology and post-Newtonian physics, but not all phenomena can be described by the linear perturbation theory. Thus, It is necessary to study exact solutions or higher order perturbations. Specifically, we study black hole (apparent) horizons and the cosmological event horizon formation in the perturbation theory. We emphasize that in the perturbative regime of the gravitational potential these horizons cannot form in the lower order. Studying the infinite plane metric, we show that to capture the cosmological constant effect we need at least a second order expansion.

Starting with the relativistic Boltzmann equation for a system of particles defined by a distribution function, we have derived the virial relation for a spherical structure within an expanding background in the context of general relativity. This generalized form of the virial relation is then applied to the static case of a spherically symmetric structure to see the difference in the simplest case to the Newtonian relation. A relativistic Mass-Temperature relation for this simple case is also derived which can be applied to compact objects in astrophysics. Our general virial relation is then applied to the non-static case of a structure within an expanding universe where an extra term, usually missed in studies of structures in the presence of the dark energy, appears.

We have constructed a spherically symmetric structure model in a cosmological background filled with perfect fluid with non-vanishing pressure as an exact solution of Einstein equations using the Lema\^{i}tre solution. To study its local and quasi-local characteristics including the novel features of its central black hole, we have suggested an algorithm to integrate the equations numerically. The result shows intriguing effects of the pressure inside the structure. The evolution of the central black hole within the FRW universe, its decoupling from the expanding parts of the model, the structure of its space-like apparent horizon, the limiting case of the dynamical horizon tending to a slowly evolving horizon, and the decreasing mass in-fall to the black hole is also studied. The cosmological redshift of a light emitted from the cosmological structure to an observer in the FRW background is also calculated. This cosmological redshift includes local and cosmic part which are explicitly separated. We have also formulated a modified NFW density profile for a structure to match the exact solution conditions.

Using a general relativistic exact model for spherical structures in a cosmological background, we have put forward an algorithm to calculate the test particle geodesics within such cosmological structures in order to obtain the velocity profile of stars or galaxies. The rotation curve thus obtained is based on a density profile and is independent of any mass definition which is not unique in general relativity. It is then shown that this general relativistic rotation curves for a toy model and a NFW density profile are almost identical to the corresponding Newtonian one, although the general relativistic masses may be quite different.

We use an exact general relativistic model structure within an FRW cosmological background based on a LTB metric to study the gravitational lensing by a cosmological and dynamical structure. Using different density profiles for the model structure, the deviation angle and the time delay through the gravitational lensing has been studied by solving the geodesic equations. The results of these exact calculations have been compared to the thin lens approximation. We have shown that the result for the thin lens approximation based on a modified NFW density profile with a void before going over to the FRW background matches very well with the exact general relativistic calculations. However, the thin lens approximation based on a normal NFW profile does differ from the exact relativistic calculation. The difference is more the less compact the structure is. We have also looked at the impact of our calculation on the observational interpretation of arcs in the case of strong lensing and also the reduced shear in the case of weak lensing. No significant difference has been seen in the data available.

We have constructed a spherically symmetric structure model in a cosmological background filled with perfect fluid with non-vanishing pressure and studied its quasi-local characteristics. This is done by using the Lema\^{i}tre solution of the Einstein equations and suggesting an algorithm to integrate it numerically. The result shows intriguing effects of the pressure inside the structure. The evolution of the central black hole within the FRW universe, its decoupling from the expanding parts of the model, the structure of its space-like apparent horizon, the limiting case of the dynamical horizon tending to a slowly evolving horizon, and the decreasing mass in-fall to the black hole is also studied. The quasi-local features of this cosmological black hole may not be inferred from the weak field approximation although the gravity outside the structure is very weak.

We have used an exact general relativistic model structure within a FRW cosmological background based on a LTB metric to study the gravitational lensing of a cosmological structure. The integration of the geodesic equations turned out to be a delicate task. We realized that the use of the rank 8(7) and 10(11) Runge-Kutta numerical method leads to a numerical effect and is therefore unreliable. The so-called semi-implicit Rosenbrock method, however, turned out to be a viable integration method for our problem. The deviation angle calculated by the integration of the geodesic equations for different density profiles of the model structure was then compared to those of the corresponding thin lens approximation. Using the familiar NFW density profile, it is shown that independent of the truncation details the thin lens approximation differ substantially from the exact relativistic calculation. The difference in the deflection angle for different impact parameters may be up to about 30 percent. However, using the modified NFW density profile with a void before going over to the FRW background, as required by an exact general relativistic model, the thin lens approximation coincides almost exactly with the general relativistic calculation.

Using a cosmological black hole model proposed recently, we have calculated the quasi-local mass of a collapsing structure within a cosmological setting due to different definitions put forward in the last decades to see how similar or different they are. It has been shown that the mass within the horizon follows the familiar Brown-York behavior. It increases, however, outside the horizon again after a short decrease, in contrast to the Schwarzschild case. Further away, near the void, outside the collapsed region, and where the density reaches the background minimum, all the mass definitions roughly coincide. They differ, however, substantially far from it. Generically, we are faced with three different Brown-York mass maxima: near the horizon, around the void between the overdensity region and the background, and another at cosmological distances corresponding to the cosmological horizon. While the latter two maxima are always present, the horizon mass maxima is absent before the onset of the central singularity.

Application of concepts like black hole and event horizon in cosmological context are not trivial, as has been shown in the last decade. We introduce special solutions of the LTB family representing collapsing over-dense regions extending to an expanding closed, open, or flat FRW model asymptotically. We study the dynamics of the collapsing region, and its density profile. The question of the strength of the central singularity and its nakedness, as well as the existence of an apparent horizon and an event horizon is dealt with in detail. Differences to the Schwarzschild black hole are addressed.

We study the volume averaging of inhomogeneous metrics within GR and discuss its shortcomings such as gauge dependence, singular behavior as a result of caustics, and causality violations. To remedy these shortcomings, we suggest some modifications to this method. As a case study we focus on the inhomogeneous structured FRW model based on a flat LTB metric. The effect of averaging is then studied in terms of an effective backreaction fluid. It is shown that, contrary to the claims in the literature, the backreaction fluid behaves like a dark matter component, instead of dark energy, having a density of the order of 10^{-5} times the matter density, and most importantly, it is gauge dependent.

Concepts developed in the gravitational lensing techniques such as shear, convergence, tangential and radial arcs maybe used to see how tenable inhomogeneous models proposed to explain the acceleration of the universe models are. We study the widely discussed LTB cosmological models. It turns out that for the observer sitting at origin of a global LTB solution the shear vanishes as in the FRW models, while the value of convergence is different which may lead to observable cosmological effects. We also consider Swiss-cheese models proposed recently based on LTB with an observer sitting in the FRW part. It turns out that they have different behavior as far as the formation of radial and tangential arcs are concerned.

We present a new model universe based on the junction of FRW to flat Lemaitre-Tolman-Bondi (LTB) solutions of Einstein equations along our past light cone, bringing structures within the FRW models. The model is assumed globally to be homogeneous, i.e. the cosmological principle is valid. Local inhomogeneities within the past light cone are modeled as a flat LTB, whereas those outside the light cone are assumed to be smoothed out and represented by a FRW model. The model is singularity free, always FRW far from the observer along the past light cone, gives way to a different luminosity distance relation as for the CDM/FRW models, a negative deceleration parameter near the observer, and correct linear and non-linear density contrast. As a whole, the model behaves like a FRW model on the past light cone with a special behavior of the scale factor, Hubble and deceleration parameter, mimicking dark energy.

Matching of a LTB metric representing dust matter to a background FRW universe across a null hypersurface is studied. In general, an unrestricted matching is possible only if the background FRW is flat or open. There is in general no gravitational impulsive wave present on the null hypersurface which is shear-free and expanding. Special cases of the vanishing pressure or energy density on the hypersurface is discussed. In the case of vanishing energy momentum tensor of the null hypersurface, i.e. in the case of a null boundary, it turns out that all possible definitions of the Hubble parameter on the null hypersurface, being those of LTB or that of FRW, are equivalent, and that a flat FRW can only be joined smoothly to a flat LTB.

We present an analytically solvable nonlinear model of structure formation in a Universe with only dust. The model is an LTB solution (of General Relativity) and structures are shells of different density. We show that the luminosity distance-redshift relation has significant corrections at low redshift when the density contrast becomes nonlinear. A minimal effect is a correction in apparent magnitudes of order 0.15. We discuss different possibilities that could further enhance this effect and mimick Dark Energy.

We present the generalized Friedmann equations describing the cosmological evolution of a finite thick brane immeresed in a five-dimensional Schwarzschild Anti-de Sitter spacetime. A linear term in the density in addition to a quatratic one arises in the Friedmann equation, leading to the standard cosmological evolution at late times without introducing an ad hoc tension term for the brane. The effective four-dimensional cosmological constant is then uplifted similar to the KKLT effect and vanishes for a brane thickness equal to the AdS curvature size, up to the third order of the thickness. The four-dimensional gravitational constant is then equal to the five-dimensional one divided by the AdS curvature radius, similar to that derived by dimensional compactification. An accelerating brane cosmology may emerge at late times provided there is either a negative transverse pressure component in the brane energy-momentum tensor or the effective brane cosmological constant is positive.

We propose a model universe in the matter dominated phase described by a FRW background with local inhomogeneities, like our local patch, grown out of the primordial fluctuations. Our sub-horizon local patch consisting of different structures is approximated as an inhomogeneous cosmic fluid described by a LTB metric embedded in a background FRW universe, in which the observer could be located anywhere. Within the exact general relativistic formulation, the junction conditions for the only possible matching without a thin shell at the boundary, neglected so far in the literature, constrains the model in such a way that the luminosity distance-red shift relation mimics a FRW universe with dark energy. Therefore, the dimming of SNIa is accounted for in such a {\it structured} FRW universe. We have also calculated the exact general relativistic backreaction term and shown how it influences the global Hubble parameter and the effective density of the cosmic fluid By using an exact formulation of the general relativistic dynamics of structures in a homogeneous universe, the claim is therefore stressed that the backreaction of cosmological perturbations leads to an apparent dimming of the cosmological distances.

Within the recently proposed structured FRW model universe the averaged Einstein equations are derived. The backreaction turns out to have an interesting behavior. Its equivalent density and pressure, being proportional, are negative at early times of the dark ages of the universe, and change sign near our present time in our local patch. In addition to explaining the observed dimming of the SNIa it leads to new effects for small cosmic redshifts and also to the difference between the local and global Hubble parameter. Interpreting the backreaction in the FRW-{\it picture}, it is equivalent to a time dependent dark energy with $w = -1$.

Concepts like peculiar velocity, gravitational force, and power spectrum and their interrelationships are of utmost importance in the theories of structure formation. The observational implementation of these concepts is usually based on the Newtonian hydrodynamic equations, but used up to scales where general relativistic effects come in. Using a perturbation of FRW metric in harmonic gauge, we show that the relativistic effects reduce to light cone effects including the expansion of the universe. Within the Newtonian gravitation, the linear perturbation theory of large scale structure formation predicts the peculiar velocity field to be directly proportional to gravitational force due to the matter distribution. The corresponding relation between peculiar velocity field and density contrast has been given by Peebles. Using the general relativistic perturbation we have developed, this familiar relation is modified by doing the calculation on the light cone in contrast to the usual procedure of taking a space-like slice defined at a definite time. The velocity and density--spectrum are compared to the familiar Newtonian expressions. In particular, the relativistic $\beta$--value obtained is reduced and leads to an increased bias factor or a decreased expected amount of the dark matter in a cluster.

Applying the distributional formalism to study the dynamics of thin shells in general relativity, we regain the junction equations for matching of two spherically symmetric spacetimes separated by a singular hypersurface. In particular, we have shown how to define and insert the relevant sign functions in the junction equations corresponding to the signs of the extrinsic curvature tensor occurred in the Darmois-Israel method.

Time variation of Newtonian gravitational constant, $G$, is studied in the model universe with variable space dimension proposed recently. Using the Lagrangian formulation of these models, we find the effective gravitational constant as a function of time. To compare it with observational data, a test theory for the time variation of $G$ is formulated. We have assumed a power law behavior of the time variation of $G$ where the exponent $\beta$ is itself time dependent. Within this test theory we are able to restrict the free parameter of the theories under consideration and give upper bounds for the space dimension at the Planck era. The time variation of $G$ at earlier times, such as the time of nucleosynthesis is also predicted which express the needs to look for related observational data.

Assuming the space dimension is not constant, but varies with the expansion of the Universe, a Lagrangian formulation for a toy model Universe is given. There is a free paremeter in the theory, $C$, with which we can fix the dimension of space at the Planck time. The standard FRW model corresponds to the limiting case $C \to +\infty$. We study the Wheeler-De Witt equation and the wave function of the Universe and the probability density in our model Universe. In the limit of constant space dimension, our wave function approaches to the Hartle-Hawking wave function or to a modified Linde wave function.

A systematic rotation of the plane of polarization of electromagnetic radiation propagating over cosmological distances has been claimed to be detected by Nodland and Ralston.Taking the finite rotating universe of Oszvath and Sch\"ucking as a toy model for the actual universe,we calculate the rotation of the plane of polarization of electromagnetic waves in this universe.It turns out that the observed data are consistent with the Oszvath-Sch\"ucking parameter $k=0.39$.

We propose a model universe, in which the dimension of the space is a continuous variable, which can take any real positive number. The dynamics leads to a model in which the universe has no singularity. The difference between our model and the standard Friedman-Robertson-Walker models become effective for times much before the presently accepted age of the universe.