Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
Permanent URI for this collectionhttps://hdl.handle.net/11147/7148
Browse
4 results
Search Results
Article Robust Scheduling of Crude Oil Farming and Processing Under Uncertainty(Elsevier, 2026) Yalcin, Damla; Sildir, HasanThe sulphur content in crude oil has a significant impact on refinery operations, influencing the feasibility of crude blending, the distribution of product yields, and overall economic performance. Variations in sulphur content introduce uncertainty in the short-term scheduling of crude oil loading, blending, and distillation processes. This study introduces a scenario-based stochastic optimization framework in which sulphur uncertainty is treated as a central modeling element, represented through a regression-based relationship with specific gravity (SG). The approach systematically propagates uncertainty through blending decisions, crude distillation unit (CDU) feed composition, and product yields. The problem is modeled as a mixed-integer quadratically constrained programming (MIQCP) formulation within a continuous-time scheduling framework, enabling the simultaneous optimization of timing, blending, and processing strategies. The results indicate that increased sulphur uncertainty adversely affects the distribution of yields for nine end-products, resulting in profit losses. These findings underscore the importance of explicitly managing compositional uncertainty and provide insights into cost-performance trade-offs in refinery scheduling.Article Towards Facile Deep Learning Architectures for Chemical Processes: Simultaneous Pseudo-Global Training and Economic Synthesis(Institution of Chemical Engineers, 2025) Sildir, Hasan; Yalcin, Damla; Tuncer, Basak; Deliismail, Ozgun; Leblebici, Mumin EnisChemical process data is usually not directly valorized in pure machine learning predictive models due to limited data availability. This limitation often caused from high sensor costs, data variety, and veracity issues. In response, this study proposes a novel formulation based on mixed-integer linear programming (MILP), called Approximated Deep Learning (ADL), to overcome these limitations and enable accurate modeling under data scarcity. The ADL simultaneously performs input selection, outlier filtering, and training of deep learning architectures within a single-level optimization problem. The method approximates the nonlinear and nonconvex components of traditional deep learning models in the mixed-integer domain through sophisticated reformulations, achieving a pseudo-global solution. A key feature of ADL is the integration of sensor pricing as a regularization mechanism, which promotes cost-efficient soft sensor design without compromising predictive performance. The proposed framework is validated on a publicly available bubble column dataset and benchmarked against four conventional deep learning methods. Results show that ADL achieves superior test accuracy with more than 50% reduction in input space, drastically reducing sensor cost. Furthermore, the optimized architecture is a high-quality initial guess for transfer learning on larger datasets. Overall, the method offers a practical and economically viable solution for data-driven chemical process modeling.Article Citation - WoS: 1Citation - Scopus: 1Storage Tank Farming Planning Under Equipment and Port Operational Costs Through Mixed Integer Quadratically Constrained Programming(Elsevier, 2025) Yalcin, Damla; Deliismail, Ozgun; Tuncer, Basak; Sildir, HasanThe study contributes a new method for managing crude oil tank farms, focusing on scheduling and optimizing storage tanks using mathematical modeling. The short-term continuous-time scheduling model reduces tank requirements and performs selection with convenient capacities. The nonconvex mixed-integer quadratically constrained programming (MIQCP) model is used to account for tank farm scheduling dynamics. It focuses on the integration of ships, storage tanks, charging tanks, and crude oil distillation units. The study examines 8 cases focusing on oil supply, arrival times, prices, and maximum flow rate constraints to show the impact of real-world volatility. By incorporating process intensification principles, the mathematical model emphasizes the importance of optimizing storage tank usage to minimize port operational costs of crude oils.Article Citation - WoS: 5Citation - Scopus: 5Automated Deep Learning Model Development Based on Weight Sensitivity and Model Selection Statistics(Pergamon-elsevier Science Ltd, 2025) Yalcin, Damla; Deliismail, Ozgun; Tuncer, Basak; Boy, Onur Can; Bayar, Ibrahim; Kayar, Gizem; Sildir, HasanCurrent sustainable production and consumption processes call for technological integration with the realm of computational modeling especially in the form of sophisticated data-driven architectures. Advanced mathematical formulations are essential for deep learning approach to account for revealing patterns under nonlinear and complex interactions to enable better prediction capabilities for subsequent optimization and control tasks. Bayesian Information Criterion and Akaike Information Criterion are introduced as additional constraints to a mixed-integer training problem which employs a parameter sensitivity related objective function, unlike traditional methods which minimize the training error under fixed architecture. The resulting comprehensive optimization formulation is flexible as a simultaneous approach is introduced through algorithmic differentiation to benefit from advanced solvers to handle computational challenges and theoretical issues. Proposed formulation delivers 40% reduction, in architecture with high accuracy. The performance of the approach is compared to fully connected traditional methods on two different case studies from large scale chemical plants.