Machine-Learning-Based Multi-Surrogate-Assisted Joint Optimization for Hydraulic Fracturing Design and Production Control

Machine learning-based model for evaluating multi-perforation hydraulic fracturing and analysis of fracturing mechanisms

An efficient surrogate-assisted hybrid optimization framework for polymer flooding management

Reservoir injection-production optimization is critical for enhancing oil recovery in high-water-cut mature fields. Polymer flooding improves development efficiency by increasing fluid viscosity and controlling mobility, yet its optimization faces significant challenges: conventional numerical simulation is computationally prohibitive, the incorporation of slug parameters introduces high-dimensional mixed-variable modeling difficulties, and single-stage optimization frameworks frequently converge to local optima in heterogeneous reservoirs. To address these issues, this study proposes a novel Two-Stage Collaborative Optimization (TSCO) framework. First, a hybrid-variable modeling approach is developed, employing binary encoding for slug timing and a Gower-distance-enhanced radial basis function network (RBFNmv) surrogate to map discrete–continuous relationships. Second, a two-stage surrogate-driven architecture is introduced: a global phase uses a dual-mode surrogate search to identify promising regions, followed by a local refinement phase for continuous parameter adjustment, with an adaptive switching mechanism balancing exploration and exploitation. Validation on two benchmark reservoir models demonstrates that TSCO achieves a significant improvement over baseline methods, delivering a net present value about 5% higher than Differential Evolution and converging to high-quality solutions about 22% faster than a conventional surrogate-assisted evolutionary algorithm. The RBFNmv surrogate model shows high predictive accuracy, explaining over 94% of the NPV variance. The key novelty lies in the integrated framework that dynamically handles mixed variables, adaptively switches search stages, and efficiently combines global exploration with local intensification—advancing beyond static, single-strategy surrogate-assisted methods. This work provides a systematic, efficient, and scalable solution for polymer flooding management, offering both methodological innovation and practical insights for field application.

A CFD-driven surrogate-based machine learning framework for multi-criteria quality optimization of family-mold processes

This study presents a machine learning-driven framework for the constrained multi-criteria optimization of family-mold processes, where the simultaneous production of geometrically dissimilar parts creates complex flow imbalances. Dimensional deformation (warpage) and shear stress are identified as critical quality characteristics affecting the functional performance and structural integrity of molded products, while product weight is treated as a practical manufacturing constraint to ensure material efficiency. Preliminary experimental trials are conducted to verify process feasibility and to determine realistic operating ranges for key injection molding parameters. Based on these ranges, a face-centered central composite design (FCCCD) is constructed, and high-fidelity computational fluid dynamics (CFD) data are generated using Moldex3D analysis. A Kriging-based surrogate model is then developed to represent the complex and nonlinear relationships between process parameters and quality responses. The surrogate model is integrated with the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to perform constrained multi-criteria optimization, where warpage and shear stress are simultaneously minimized under a prescribed weight constraint. The optimization process yields Pareto-optimal process parameter combinations that effectively balance dimensional accuracy and stress reduction while satisfying material usage requirements. Selected Pareto-optimal solutions are validated through detailed CFD simulations, yielding average relative errors of 1.56 % and 3.58 % for warpage and shear stress, respectively, demonstrating the predictive accuracy and robustness of the proposed framework. Overall, this study establishes an effective data-driven optimization strategy for quality-oriented design and decision support in intelligent family-mold processes.

Efficient Geothermal Well-Control Optimization via Diffusion-Surrogate Reinforcement Learning

Real-time decision-making for enhanced geothermal systems (EGS) is challenging because long-term production periods involve high-dimensional control spaces and a large number of time-consuming high-fidelity hydrothermal simulations. Reinforcement learning provides a natural framework for state-dependent sequential control, but direct policy training with numerical simulators is computationally expensive. To address this issue, we propose a diffusion-surrogate guided reinforcement learning framework for long-horizon EGS well-control optimization. The reservoir temperature and pressure fields are used as system states, while injection rates are selected as control actions. A learned surrogate environment is constructed using conditional diffusion models to predict the evolution of reservoir temperature and pressure fields and a separate reward model to estimate the corresponding economic return. The surrogate environment is then integrated with Proximal Policy Optimization (PPO) for efficient policy training. Experiments on a fractured EGS benchmark show that the diffusion surrogate can accurately reproduce reservoir-state evolution over multiple control stages. The resulting surrogate-assisted PPO policy achieves competitive well-control performance compared with direct simulator-based PPO and existing optimization methods, while substantially reducing the dependence on expensive high-fidelity simulations. These results demonstrate the potential of diffusion-based surrogate environments for efficient reinforcement learning in geothermal well-control optimization.

Machine-Learning-Assisted Prediction of Port-Flow Distribution and Multi-Objective Parametric Optimization for Navigation Lock Manifolds

Navigation lock manifolds are key components of filling-and-emptying systems, and port-flow distribution affects chamber flow stability and filling efficiency. Under unsteady filling conditions, port-flow distribution is governed by discharge variation and manifold geometry, making rapid prediction and engineering-constrained screening challenging. This study develops a surrogate-assisted prediction and Pareto-screening framework for a large-scale navigation lock manifold. Three-dimensional computational fluid dynamics (CFD) simulations were used to examine unsteady port-flow evolution. The peak-flow condition was selected as a representative control condition, and the flow non-uniformity coefficient α and system resistance coefficient ξ were used as performance indicators. Based on 243 parametric CFD samples and 144 independent external test samples, artificial neural network (ANN), Gaussian process regression (GPR), and support vector regression (SVR) models were evaluated. ANN performed best, with independent-test R2 values of 0.9999 and 0.9928 for α and ξ. Feature-attribution analysis identified port width, culvert height, and port number as dominant variables. Pareto screening within a predefined engineering design space identified representative candidates with CFD verification errors below 1.1%. The TOPSIS-based candidate reduced ξ by 32.2% while maintaining α nearly unchanged.

A Forecast-Informed Rule-Based Control Framework for Optimizing the Hydraulic Performance of Bioretention Systems

Bioretention systems are widely implemented to mitigate urban runoff; however, conventional passive designs lack the ability to adapt discharge behavior to evolving rainfall conditions, often resulting in premature drainage and inefficient utilization of available storage. This study develops and evaluates a forecast-informed, rule-based control framework for optimizing the long-term hydraulic performance of bioretention systems. The proposed approach integrates continuous valve actuation with a kernel-based rainfall forecast indicator embedded directly within a process-based hydraulic model implemented in OpenHydroQual. Control parameters governing forecast interpretation and valve responsiveness are optimized offline using a genetic algorithm over multiyear simulations. System performance is evaluated using three complementary metrics: exceedance flow at a probability level of p = 0.001 , total discharged volume, and a residence-time objective, defined as the outlet load of a Laplace-transformed age variable. Simulation-based evaluation using a field-calibrated model of a dual bioretention system demonstrated that exceedance-only optimization reduces extreme discharge from 1.99 × 10 2 to 2.86 × 10 1    m 3 / day (approximately 86%) relative to passive operation. Multiobjective optimization revealed a well-defined Pareto trade-off between exceedance suppression and residence-time enhancement. Increasing the residence-time weight yielded a 52% reduction in the transformed-age load while increasing exceedance by less than 20% relative to the exceedance-only solution. Notably, moderate weighting achieved approximately 40% improvement in residence time with less than 3% increase in exceedance, indicating a region of hydraulically efficient operating conditions. Results demonstrate that forecast-informed rule-based control can substantially improve both extreme-flow mitigation and storage persistence without requiring online optimization or computationally intensive model predictive control. The proposed framework provides a transparent, computationally lightweight, and physically interpretable approach for enhancing the operational performance of existing bioretention infrastructure under long-term, variable hydrologic conditions.