Lizeth Torres

Effective decision-making in reservoir management requires accurate insight into interwell fluid-flow communication and the dynamic response of fluids in fractured reservoirs. This study builds upon prior Visibility Graph (VG) analysis and Multiplex Network (MN) methodologies to infer potential interwell communication patterns associated with subsurface fluid flow connectivity. We leverage VG-derived adjacency matrices in conjunction with production and injection time series, treating them as dynamic reservoir responses and transforming them into image representations using the Gramian Angular Field (GAF) method. These GAF images then serve as input to a Convolutional Neural Network (CNN) for classifying zones associated with different levels of inferred interwell fluid-flow connectivity. The result is a dual-path framework that combines structural insights from VG with learned spatiotemporal patterns from production-injection dynamics, enabling the generation of spatial diagnostic maps for reservoir planning. These maps provide a proxy interpretation of possible preferential communication pathways associated with production-injection response. The proposed method supports practical decision-making by providing interpretable spatial representations of reservoir connectivity and production-injection interaction patterns. Comparison of the resulting metrics demonstrates the potential of the method to support reservoir diagnostics and identify zones where connectivity-related behavior may require further engineering evaluation. Furthermore, we assess two GAF-based input strategies, the impact of data augmentation, and the resulting spatial classification of interwell connectivity zones.

The photoacoustic (PA) effect is an emerging technique for characterizing the structural and compositional features of materials by combining light absorption with the generation of pressure waves. In practice, it is only possible to measure the pressure wave at the sample surface using an ultrasound transducer, namely a PA signal. However, reconstructing the initial pressure is an inverse problem with a non-trivial solution. This work introduces a dynamic state estimator (DSE) that virtually measures wave propagation to compute the initial pressure profile from boundary measurements. The proposal is based on a dynamical model that describes the spatiotemporal evolution of a PA signal, where the states represent the pressure waves within the sample. Through the DSE, the method provides a smooth reconstruction of the initial pressure in a straightforward yet robust manner. The state estimator is tested in both numerical and real experiments, outperforming traditional model-based algorithms and yielding results comparable to the well-known time-reversal algorithm in signal-to-noise ratio, contrast, and computational speed. The results suggest that the state estimator represents a technological advancement in overcoming the inherent ill-posedness of the PA inverse problem in a wide range of applications.

This review summarizes the latest research concerning the horizontal flow of two-phase mixtures with viscosities ranging from 0.2 Pa·s to 6.4 × 10⁴ Pa·s. Although our survey is concerned with Newtonian fluids, a short section is included to briefly discuss certain rheological aspects that should be generally considered. In contrast with previous work reporting on the progress in specific domains (e.g., in the oil and gas, chemical, or geophysical contexts), we seek to provide a comprehensive overview of the methods and results used in different contexts. Accordingly, the scope is widened to encompass a broader range of industrial applications and naturally occurring flows. The interest in high-viscosity flows is motivated by the operational challenges occurring in certain systems, most notably in the oil and gas industry, where the production of heavy and extra-heavy crude oils reduces the margins for a safe and efficient operation. Furthermore, this review underlines the cross-field analogies appearing in a broad range of scales and applications. It emphasizes the fundamental role of viscosity in determining the flow patterns, as experimental evidence suggests that the transition boundaries are largely altered at higher viscosities. Some gaps that could be addressed in future work are briefly discussed.

Principalement, ce travail présente l’application d’observateurs non linéairespour la détection de fuites (uniques, séquentielles et simultanées) dans des canalisationssous pression. Les observateurs présentés ici ont été conçus à partir d’uneversion discrète des équations du coup de bélier, qui a été obtenue en utilisant laméthode des différences finies et en prenant comme alternative la méthode de collocationorthogonale. Les modèles discrets ainsi que certains observateurs ont étévalidés par une série d’expériences effectuées dans des canalisations d’essai. D’autrepart, une nouvelle version d’observateurs à grand gain pour des systèmes non uniformémentobservables a été développée. Elle a été utilisée pour la détection de fuitesainsi que pour la synchronisation de systèmes chaotiques avec des paramètres inconnus.Des résultats de convergence, expérimentaux et en simulation sont exposésdans ce mémoire.

This paper proposes a model-based methodology to estimate multiple nodal demands by using only pressure and flow rate measurements, which should be recorded at the inlet of the distribution system. The algorithm is based on an array of multiple extended Kalman filters (EKFs) in a cascade configuration. Each EKF functions as an unknown input observer and focuses on a section of the pipeline. The pipeline model used to design the filters is an adaptation of the well-known rigid water column model. Simulation and experimental tests on standardized pipeline systems are presented to demonstrate the proposed method’s effectiveness. Finally, for the case of the experimental validation, both steady-state and variable input conditions were considered.

One of the main means of monitoring aquatic bodies are buoys, most of which are fixed, so they are not flexible enough to perform tasks such as: cleaning; rescue; exploration and surveillance. Therefore, this work is dedicated to the mathematical modeling and design of control schemes for a compact buoy with omnidirectional configuration to move between two desired points in a given time; this configuration allows the buoy to converge accurately and quickly to desired positions and routes, however, disturbances due to waves must be considered, which directly affect the movement of the buoy. These disturbances can be compensated by a robust motion controller. Thus, the dynamic model of the buoy is obtained and two control schemes are developed: Proportional-Derivative-Plus controller and Backstepping controller. These controllers base their design on a Lyapunov candidate function, which guarantees the stability of the system; however, they depend on the parameters of the dynamic model. Therefore, the dynamic buoy model is obtained considering environmental disturbances such as wind and waves. The mathematical model and the control algorithms are simulated and compared in MATLAB/Simulink. The comparison is made considering the presence of environmental disturbances from waves and wind. In addition, the evaluation was carried out using performance indicators such as variation in control efforts and those based on error: integral error, absolute integral error, and mean square error.