Search

The present study applies a Deep Reinforcement Learning (DRL) algorithm to Active Flow Control (AFC) of a two-dimensional flow around a confined square cylinder. Specifically, the Soft Actor-Critic (SAC) algorithm is employed to modulate the flow of a pair of synthetic jets placed on the upper and lower surfaces of the confined squared cylinder in flow configurations characterized by Re of 100, 200, 300, and 400. The investigation starts with an analysis of the baseline flow in the absence of active control. It is observed that at Re = 100 and Re = 200, the vortex shedding exhibits mono-frequency characteristics. Conversely, at Re = 300 and Re = 400, the vortex shedding is dominated by multiple frequencies, which is indicative of more complex flow features. With the application of the SAC algorithm, we demonstrate the capability of DRL-based control in effectively suppressing vortex shedding, while significantly diminishing drag and fluctuations in lift. Quantitatively, the data-driven active control strategy results in a drag reduction of approximately 14.4%, 26.4%, 38.9%, and 47.0% for Re = 100, 200, 300, and 400, respectively. To understand the underlying control mechanism, we also present detailed flow field comparisons, which showcase the adaptability of DRL in devising distinct control strategies tailored to the dynamic conditions at varying Re. These findings substantiate the ability of DRL to control chaotic, multi-frequency dominated vortex shedding phenomena, underscoring the robustness of DRL in complex AFC problems. [ABSTRACT FROM AUTHOR]

Deep reinforcement learning (DRL) has emerged as a promising approach for handling highly dynamic and nonlinear active flow control (AFC) problems. However, the computational cost associated with training DRL models presents a significant performance bottleneck. To address this challenge and enable efficient scaling on high-performance computing architectures, this study focuses on optimizing DRL-based algorithms in parallel settings. We validate an existing state-of-the-art DRL framework used for AFC problems and discuss its efficiency bottlenecks. Subsequently, by deconstructing the overall framework and conducting extensive scalability benchmarks for individual components, we investigate various hybrid parallelization configurations and propose efficient parallelization strategies. Moreover, we refine input/output (I/O) operations in multi-environment DRL training to tackle critical overhead associated with data movement. Finally, we demonstrate the optimized framework for a typical AFC problem where near-linear scaling can be obtained for the overall framework. We achieve a significant boost in parallel efficiency from around 49% to approximately 78%, and the training process is accelerated by approximately 47 times using 60 central processing unit (CPU) cores. These findings are expected to provide valuable insight for further advancements in DRL-based AFC studies. [ABSTRACT FROM AUTHOR]

This study investigates the utilization of a trapezoidal cavity with a corrugated bottom wall containing a Carreau hybrid nanofluid composed of water, aluminum, and copper nanoparticles. A heated, rotating cylinder is placed at the center of the enclosure in the presence of an external magnetic field. The impact of Forchheimer and Brinkman porous medium models on the hybrid nanofluid is examined. Three different inlet–outlet placement configurations are considered to investigate their influence on heat transfer. The governing equations for fluid flow and heat transfer are solved numerically. Through simulations, a range of flow-controlling variables is systematically adjusted, including the Darcy number, Reynolds number, Hartmann number, nanoparticle volume fraction, undulation on the hot bottom wall, power law index, and rotational speed of the inner heated cylinder. The results demonstrate that the hybrid nanofluid and rotating cylinder significantly enhance heat transfer within the trapezoidal cavity. Higher values of the Darcy number, Reynolds number, and nanoparticle volume fraction lead to increased heat transfer rates. The placement configuration of the inlet and outlet ports also affects heat transfer performance, with the bottom-top configuration yielding the best results. Furthermore, a comparative analysis of flow profiles and heat distribution is conducted using the multiple expression programing technique. The proposed model accurately predicts the flow and heat transfer characteristics in the trapezoidal cavity, as validated through comparison with provided data sets. [ABSTRACT FROM AUTHOR]

Nanoemulsions have garnered great attention as a chemical additive for enhanced oil recovery (EOR) technology worldwide due to their small size and unique physicochemical properties. Herein, this work introduces a novel nonionic surfactant-stabilized nanoemulsion prepared using a low-energy method for EOR projects in the Chang 8 tight reservoir. Spontaneous imbibition experiments were conducted in oil-saturated cores, and secondary imbibition experiments were carried out in water-bearing cores. Additionally, core flooding experiments were performed to study the EOR performance of both nanoemulsion and complex surfactant. Finally, the EOR mechanisms were thoroughly investigated, specifically focusing on oil-washing capacity, reduced interfacial tension, and wettability alteration. The designed nanoemulsion, with droplets averaging 10.5 ± 0.7 nm in synthetic formation brine at ambient temperature, exhibits excellent thermal and long-term stability even under reservoir temperature conditions. The oil recovery efficiency of the nanoemulsion system in oil-saturated cores through spontaneous imbibition was 31.28%, representing a 6.42% improvement over the complex surfactant solution. In the secondary imbibition experiments, the nanoemulsion mobilized 8.53% of the residual oil, compared to 4.11% for the complex surfactant. Core flooding experiments revealed that the nanoemulsion system achieved a total oil recovery of 62.69%, outperforming the complex surfactant system by 8.68% during the chemical flooding stage and 1.82% during the subsequent brine flooding stage. The EOR mechanisms of nanoemulsion in tight reservoirs can be attributed to four aspects: reduction of oil/water interfacial tension, wettability alteration, miscibility with crude oil, and increased sweep volume. The generation of the Marangoni effect and the adsorption and diffusion characteristics are essential differences between the nanoemulsion system and the complex surfactant system. The presented findings in this paper could aid in promoting the large-scale use of nanoemulsions in Chang 8 tight reservoirs. [ABSTRACT FROM AUTHOR]

This article investigates the fluid dynamics and heat transfer properties in a trapezoidal enclosure containing a heated cylindrical object. It involves the interaction of multiple physical processes such as the magnetic field, thermal radiation, porous materials, and aqueous copper oxide nanoparticles. The governing partial differential equations are analyzed numerically through the continuous Galerkin finite element algorithm. The analysis takes into account various physical parameter factors, including the Richardson number (0 – 5) , the Hartmann number (5 − 40) , the Darcy number (0.001 − 0.1) , thermal radiation parameter (0.5 − 2) , and nanoparticle volume concentration (0.01 − 0.1). The physical mechanism of thermal and mass transfer in the enclosure caused by various factors is fully explored. In addition, the multiple expression programming (MEP) technique is implemented to report a comparative analysis of flow profiles and thermal distribution. The findings demonstrated that at low Ri, the primary flow within the cavity is driven by the shear friction generated by the moving walls. The growing importance of radiative heat transfer reduces the effectiveness of convective heat transfer, resulting in a decline in the average Nusselt number with R. The heat transfer rate rises up to 27.7% as ϕ augments; however, its value declines by 9.37% against Ha. The expected results obtained by the MEP approach are very consistent with the numerical ones. There is no doubt that the new MEP concept provides a valuable tool for researchers to predict the heat transfer behavior of any data set in cavities of different shapes. It is expected to provide new idea for the development of efficient cooling systems and the improvement of energy efficiency in various engineering applications. [ABSTRACT FROM AUTHOR]

This paper considers storm surge prediction using a neural network and considering multiple physical characteristics. Based on the factors that influence storm surges and historical observation data, we divide the input to the neural network into time features extracted from the prediction target and the auxiliary features that affect storm surges, and construct a feature gate within multiple recurrent neural network (RNN) cells. Historical hurricane data are used to assess the effectiveness and accuracy of the proposed model. Comparative analysis against a long short-term memory (LSTM) storm surge prediction model is conducted to verify the prediction performance of the proposed method. The comparison results show that the multi-RNN model is superior to the LSTM model in terms of four evaluation metrics and for all lead times. In particular, the multi-RNN model accurately predicts the maximum storm surge water level, and the prediction results are more consistent with the rise and fall of the water. A comparison of the storm surge forecasts using inputs from different time intervals under different evaluation indices confirms the generalization and stability of our proposed model. The experiments of storm surge prediction at six stations further confirm the wide applicability of the model. [ABSTRACT FROM AUTHOR]

The encoder–decoder LSTM (long short term memory) recurrent neural network is proposed to predict storm surge in Florida. Two types of hurricanes with six events are collected for training and testing. The previously observed meteorological data including the storm surge, the wind speed, the wind gust, the barometric pressure, and the air temperature are chosen as the inputs of model, while the future storm surge is designated as the output. The predicted results for 1, 3, 6, 9, and 12 h-lead time are obtained. Four indices are introduced to evaluate the accuracy and stability of the proposed model. Through comparing the predicted results by both models with the observed data, it is found that the encoder–decoder LSTM approach is more accurate than the convolutional neural network model for all considered cases. Furthermore, we notice that the combinations of storm surge, wind speed, wind gust, barometric pressure, and air temperature, and storm surge and wind speed give the best prediction for the first and the second types of hurricanes, respectively. Our work suggests that the encoder–decoder LSTM model has great potential in storm surge prediction. It is expected that the accuracy of this model can be further improved by introducing more observed data and considering more physical factors. [ABSTRACT FROM AUTHOR]

After fracturing with a guar fluid system in Chang 8 oil reservoir, some wells have high oil production, and some have low production and even only produce water. Therefore, a comparative study is carried out in oil-producing and water-producing wells. The results show that: ① Organic wax adsorbed in oil-producing and water-producing well core pores accounts for 95.39% and 56.67%, respectively, making wells strong oil-wet. A certain proportion of inorganic salts and some other oil-wet organic compounds exist in water-producing wells. ② Under the reservoir conditions, the oil wettability of the water-producing well core is strong, and its original formation wettability is also oil-wet. The original wettability of the oil-producing well formation is water-wet, with oil production capacity. ③ The core residual oil saturation using the guar fluid is 24.7%–29.87%. This results in no oil phase permeability or difficulty to mobilize oil in most oil reservoirs. ④ Forced displacement can mobilize oil in large pores and has a significant effect in producing well, with mobilization efficiency of 53.49%. This article elucidates the wettability formation and crude oil mobilization mechanism of two types of wells, providing direction for the development of this type of reservoir. [ABSTRACT FROM AUTHOR]