Polymer flooding is a prominent enhanced oil recovery process that is widely recognized for its cost-effectiveness and substantial success in increasing oil production. In this study, the Buckley–Leverett mathematical model for polymer flooding was used to generate more than 163,000 samples that reflect different reservoir conditions using seven input parameters. We introduced artificial noise into the dataset to simulate real-world conditions and mitigate overfitting. Seven classic machine learning models and two neural networks were trained on this dataset to predict the oil recovery factor based on the input parameters. Among these, polynomial regression performed best with a coefficient of determination ($R^2$) of 0.909, and the dense neural network and cascade-forward neural network achieved $R^2$ scores of 0.908 and 0.906, respectively. Our analysis included permutation feature importance and metrics analysis, where key features across all models were identified, and the model’s performance was evaluated on a range of metrics. Compared with similar studies, this research uses a significantly larger and more realistic synthetic dataset that explores a broader spectrum of machine learning models. Thus, when applied to a real dataset, our methodology can aid in decision-making by identifying key parameters that enhance oil production and predicting the oil recovery factor given specific parameter values.

This work develops and systematically evaluates a physics-informed neural network (PINN) solver for the fully coupled, time-dependent Muskat–Leverett system with capillarity modeled in the pressure equation. A single shallow–wide multilayer perceptron jointly predicts wetting pressure and water saturation; physical capillary pressure regularizes the saturation front, while a small numerical diffusion term in the saturation residual acts as a training stabilizer rather than a shock-capturing device. To guarantee admissible states in stiff regimes, we introduce a saturation soft-clamping head enforcing $0<{\mathrm{S}}_{\mathrm{w}}<1$ and activate it selectively for stiff mobility ratios. Using IMPES solutions as reference, we perform a sensitivity study over network depth and width, interior collocation and boundary data density, mobility ratio, and injection pressure. Shallow-wide networks (10 layers × 50 neurons) consistently outperform deeper architectures, and increasing interior collocation points from 5000 to 50,000 reduces mean saturation error by about half, whereas additional boundary data have a much weaker effect. Accuracy is highest at an intermediate mobility ratio and improves monotonically with higher injection pressure, which sharpens yet better conditions the front. Across all regimes, pressure trains easily while saturation determines model selection, and the PINN serves as a physics-consistent surrogate for what-if studies in two-phase porous-media flow.

The advent of smart buildings owes much to the emergence of digital twins. In contemporary structures, a wealth of data is available, enabling the digital representation of buildings and facilitating improvements in energy management, particularly in heating, ventilation, and air conditioning (HVAC) systems. To effectively implement an energy management strategy within a building, a data-driven approach must accurately assess HVAC system attributes, with a focus on room temperature. Precise temperature forecasts not only enhance thermal comfort but also play a pivotal role in energy conservation. This research aims to explore data-driven methodologies and develop a model for room temperature prediction, employing machine learning algorithms in a case study of an educational building. This article details the methodology, points of interaction, and findings of this study. Relevant model parameters are identified to construct temperature prediction models using real-world data, demonstrating the efficacy of the proposed system, with an average prediction accuracy exceeding 95%. These results underscore its potential to enhance energy efficiency and thermal comfort, highlighting the use of Artificial Neural Networks (ANNs) as a pivotal component in achieving these goals.

The depletion of high-grade copper ores and the increasing prevalence of lowgrade, high-sulfur copper concentrates present significant challenges to the copper smelting industry. Traditional smelting processes struggle to maintain economic viability and comply with environmental regulations when processing these complex ores. This paper details the development of a comprehensive software solution designed to simulate the smelting process for high-sulfur, copper-poor concentrates. The software employs a detailed mathematical model to predict the yields and compositions of products, including valuable metals, during the smelting process. It integrates multiple modules, such as ore input, smelting simulation, and results presentation, providing a userfriendly platform for optimizing smelting operations. Critical parameters like ore composition, smelting temperature, and flux addition are incorporated into the model, enabling accurate predictions of matte and slag outputs. By analyzing these outputs, the software aids in optimizing metal recovery and reducing losses, ultimately enhancing the efficiency and sustainability of copper production. This tool is particularly relevant for large sulfide copper ore deposits, such as those in Kazakhstan, which have high sulfur content and low copper levels. The software's ability to simulate different processing scenarios provides valuable insights for industrial applications, supporting the development of more efficient and eco-friendly smelting technologies. The comprehensive software solution not only addresses the technical challenges of processing high-sulfur, low-copper ores but also contributes to the industry's efforts to reduce environmental impact and improve resource management. This innovation represents a significant step forward in the optimization of copper smelting operations, promoting sustainability and efficiency in the face of declining ore quality.

Physics-informed neural networks (PINNs) are difficult to train on strongly coupled, multi-objective systems: with fixed loss weights an otherwise identical run succeeds or diverges with the random seed, and a leading explanation attributes the failures to conflict between the loss-term gradients. We test that explanation on a coupled non-isothermal Buckley–Leverett displacement with a passive tracer, a three-front hot-water flood with an exact analytical solution as ground truth. The test produces an exactly verified counterexample: the conflict-free update ConFIG attains its alignment objective yet converges to a stable solution that violates the equations, while a quasi-second-order optimizer (SOAP) and per-term gradient surgery (PCGrad) train the network reliably with no loss-weight search. The most aligned update is thus the one that fails: gradient alignment is neither necessary nor sufficient for trainability, and the conflict is a diagnostic of the loss rather than its cause. The same formulation serves the forward, parametric, and inverse problems: it predicts the movable-oil recovery with a relative error of about 5%, returns the solution for any oil viscosity in a trained range from one network, and, from sparse near-injector monitoring, recovers the thermal front speed and one transport coefficient when the other is known. To our knowledge this is the first PINN solution of this coupled displacement, which is central to thermal oil recovery.