The current enhancement methods for applications such as heavy oil recovery and soil remediation are often environmentally detrimental and expensive. One alternative is acoustic excitation, an ecologically clean and inexpensive technique that can provide a new solution to enhanced recovery. Such a method has advantages over traditional solutions including eliminating bypassing effects and reducing environmental hazards. However, there are serious issues regarding the identified mechanisms including their relative contributions, scale dependency of the process, uncertainties regarding the effects of controlling factors and feasibility of some mechanisms in porous media. Therefore, the main objective of this study was to investigate whether and how acoustic excitation/stimulation can improve flow and transport during miscible and immiscible displacements in porous media.Mechanisms have been investigated during miscible and immiscible displacements by a combination of modeling and experiments. In miscible displacements, numerical models are developed across different scales to identify and utilize interactions between wave propagation and mass transfer processes. In the pore-scale, a three step approach is used to link acoustically induced particle velocity and fluid flow model. Pore-scale simulation results suggest that mixing and spreading processes in miscible displacements are impacted by pressure acoustic waves. Lower frequencies and higher acceleration amplitudes of the propagated waves increase the enhancements in dispersion and mass transfer coefficients. Based on the latter simulations, the effects of controlling parameters at pore-scale are summarized in a proposed dimensionless group of A_D for designing effective acoustically-assisted experiments in the laboratory. The results of the pore scale simulations at higher viscosity ratios also indicate that low-frequency excitation could be a promising technique for improving miscible displacements. Similar impacts were expected to be observed at large scales as well. Therefore, the interaction of fluid flow and poroelastic waves is modeled by a two-way coupling approach at macroscales. The results at a macro-scale show that mechanical stimulations could potentially be used as a hybrid solvent-based technique in heavy oil recovery. For designing such an acoustically assisted miscible system, governing dimensionless numbers are identified and used in simulation results. Higher frequencies, larger degrees of heterogeneities in rock and fluid properties, softer media and transverse waves’ propagations to the flow direction are favorable in the proposed assisted process. In immiscible displacements, pore-scale experiments are performed on microfluidic systems saturated with a dense non-aqueous phase liquid that is a common groundwater contaminant. In the experiments, two-dimensional high-quality images are processed to evaluate the break-up and mobilization of trapped ganglia at different injection rates and wettability conditions. The experimental observations are also used to develop and match a reactive pore network model that describe the remobilization and dissolution processes. Immiscible experimental results show that sonication can break up trapped ganglia in the porous system and may get them flowing under a background viscous force. In the pore network modeling, the simulation results of residual saturation reasonably matched with experimental observations. The differences between the prediction by the model and the experimental data at verification points is less than 2 % for data before and after the acoustic excitation. Based on the three-dimensional pore network simulation results, a modified capillary number is proposed as a function of acoustically induced pore pressure. This dimensionless number can be used as a predictive tool for the impact of stimulation in immiscible displacements. In the modified capillary number, the induced pore pressure is calculated by developing numerical model that links elastic, pressure acoustic and poroelastic waves at core-scale. The simulation results of this core-scale model are also correlated to the recovery enhancements to determine the required applied stress or supplied energy that causes observable changes in immiscible fluid flow behavior under laboratory conditions. In the latter study, a loss number of N_l is proposed based on the dimensionless analysis of the wave equation in fluids where the most energy dissipation occurs. The critical values of N_l are obtained by implementing a classifier machine learning algorithm on a data set from the literature.