Papers by Keyword: Optimization

Abstract: This study introduces a surrogate model-based optimization methodology to explore a wide design space of power module packages for achieving user-defined electromagnetic design objectives, such as minimizing commutation loop stray inductance, gate loop stray inductance and balancing mutual inductance. A half-bridge module with four parallel SiC devices per switch position is analyzed, incorporating 17 design variables across terminals and substrate dimensions. Using Sobol sampling, 4096 design variations were simulated in Ansys Q3D to train the surrogate model, enabling efficient gradient-based single- and multi-objective optimization. Results show that the proposed methodology significantly accelerates exploration in a wide design space and outperforms traditional expert-driven methods by identifying superior electromagnetic performance.

Abstract: High-Pressure Die Casting (HPDC) processes are often affected by complex thermo-fluid-dynamic phenomena that lead to casting defects and premature die degradation. In this study, an approach based on the Response Surface Methodology (RSM) is proposed to improve the quality of the cast part (aluminum window brackets) and extend the dies’ service life by introducing limited modification to the geometry of the die cavities.A multi-physics numerical model was initially built up to reproduce the filling and thermal behavior of the process. Infrared thermography, used to validate the numerical results, confirmed the accuracy of the model, with an average temperature error of approximately 2%. The analysis revealed that the baseline configuration (i.e. the dies’ geometry currently adopted in the industrial process) was characterized by non-negligible thermal imbalances (temperature gradients of about 50 °C and localized hot spots associated with high melt velocities), which reflected in the occurrence of flashes, metallization, and impression pad damage.New die geometries with the aim of improving the thermal uniformity while reducing the temperature gradients where investigated by varying the geometrical properties of the gating system according to a DoE-based approach. The numerical results, collected in terms of total amount of porosity in the casting critical areas, were used to train accurate metamodels that, in turns, were adopted as the starting base for a multi-objective optimization. Results from the optimization allowed to identify different scenario, each characterized by a specific geometry of the gating system able to remarkably reduce the occurrence of porosity in the cast part (up to 42% less than the current condition). The results demonstrate that the proposed methodology enables effective and sustainable optimization of HPDC processes without costly trial-and-error approaches.

Abstract: The superplastic performance of the dual-phase Ti-6242S titanium alloy makes it a good material for aerospace application to produce structural components using the advanced superplastic forming (SPF) process. The need to optimize the SPF process demands the understanding and quantifications of the influence of the different phase constituents - α and β on the global superplastic behavior. Numerical modelling has been useful to predict mechanical behavior for both one-level and multiscale approach. Multiscale approach: bottom-up (microscale to macroscale) has enabled to understand how the different microstructural parameters influence global material/structural mechanical response; which by large means the modelling approach depends on the material local properties. The identification of these local properties is non-trivial in polycrystal materials, particularly at superplastic (elevated) temperatures. We have developed a methodology that permit us to quantify the microstructural parameters of each of the constitutive phases of a polycrystal at a superplastic temperature using genetic algorithm optimization method on the data from in-situ high energy X-ray diffraction (synchrotron radiation), coupled with SEM (scanning electron microscope) and EBSD (electron backscattered diffraction). These identified local microstructural parameters were directly used in the finite strain crystal plasticity model to simulate the material global response. This approach enabled the quantification of the phase influences on global behavior with much accuracy. It was found that α phase planes have high critical resolved shearing stress (CRSS) at 730°C which is similar to its behaviours at room temperatures, while β phase slip planes have low CRSS that encourage slip shearing at low stress. However, more applied load is partitioned in β phase than in α phase, despite that β phase fraction is about 15% at 730°C. Keyword: Multiscale modelling, CPFE, optimization, HEXRD, dual-phase titanium alloy, superplasticity

Abstract: Warpage in injection-molded thin-walled box-shaped parts is primarily caused by non-uniform cooling and differential shrinkage. This study proposes a two-step, multi-objective optimization strategy to reduce part warpage by addressing both thermal and geometric factors. In the first step, the mold cooling system is optimized through a bi-objective formulation that simultaneously minimizes (i) the temperature standard deviation within the part and (ii) the total cooling channel length. The optimization is carried out using a coupled workflow involving parametric CAD modeling, Autodesk Moldflow simulations, and a genetic algorithm. The optimized cooling design reduces temperature non-uniformity by 44% compared to a conventional cooling layout. In the second step, a geometric optimization is performed through the addition of a reinforcing border, where maximum deflection and total part volume are minimized simultaneously. The combined optimization leads to a reduction in maximum warpage from 14.5 mm in the reference configuration to 2.06 mm in the final design. The results demonstrate the effectiveness of a sequential optimization approach in achieving significant warpage reduction while maintaining material and manufacturing efficiency.

Abstract: Air bending is a critical operation in the metalworking industry, where dimensional accuracy and process efficiency are essential to ensure product quality and economic viability. This work proposes an AI-driven design and optimization strategy which couples artificial intelligence, specifically artificial neural networks, with a quasi-random search algorithm for the metamodeling and optimization of the air bending process. An extensive simulation database was generated by varying geometrical, material, and process parameters, and neural-network-based metamodels were trained to predict the maximum punch force, maximum thickness reduction, and final bending angle, achieving high predictive accuracy with R² values exceeding 0.96. The metamodel was subsequently used to optimize process configurations by simultaneously minimizing the maximum punch force and the maximum thickness reduction while ensuring the target bending angle, leading on average to reductions of 46.7% in maximum force and 31.5% in thickness reduction compared to non-optimized cases. The results demonstrate that artificial intelligence provides an efficient and effective tool for the design and optimization of the bending process, significantly accelerating parameter selection while improving process quality and reducing manufacturing costs.

Abstract: This study utilizes phenanthrene as the model molecule to investigate the optimization and reusability of coal-derived carbon nanoparticles for the adsorption of polycyclic aromatic hydrocarbons (PAHs). After controlled carbonization and activation, the carbon nanoparticles were synthesized using a chemical solid synthesis method and meticulously studied to determine their surface morphology and crystallinity. One factor at a time (OFAT) was used as an optimization method for the batch adsorption studies, the parameters varied including pH, contact time, adsorbent dosage, Temperature, and initial phenanthrene concentration. The optimal circumstances for phenanthrene resulted in a high removal efficiency of up to 95.3% for phenanthrene, and 96% removal for naphthalene, hence demonstrating the material's potential for PAH remediation. Subsequent batch testing confirmed the material's efficacy in removing naphthalene and phenanthrene. Furthermore, reusability studies conducted over five adsorption-desorption cycles demonstrated minimal decline in removal efficiency for Naphthalene by 10%, with a difference between the 1^st and 5^th run. hence showing robust regeneration capability and operational stability. But it shows a high decline in removal efficiency for phenanthrene. The results demonstrate the efficacy and sustainability of coal-derived carbon nanoparticles as a cost-effective adsorbent for applications addressing PAH contamination in water.

Abstract: Heating networks are a crucial part of modern urban infrastructure, delivering heat to residential, commercial, and industrial consumers efficiently and reliably. Ensuring their safe and continuous operation is essential for maintaining comfort and supporting daily urban life. The integration of digital twins has become increasingly important, as they allow operators to monitor network behavior in real time, predict potential failures, and implement corrective actions promptly. By reducing response times and minimizing the frequency of accidents, digital twins help ensure a stable and uninterrupted heat supply. For a digital twin to be effective, it must be based on accurate numerical models that capture fluid flow, heat transfer, and pressure distribution throughout the network. Proper design and modeling enable efficient use of resources, including pumping power and pipe sizing, while reducing energy waste and operational costs. This study presents a comprehensive approach to optimizing heating networks. Control variables such as pipe diameters, pump pressure, and the settings of bypass and radiator valves for each consumer are defined. A constraint aggregation function ensures that no consumer experiences freezing, while the objective is to minimize both the initial installation costs and long-term operational expenses. Advanced numerical solvers were used to perform the calculations, enabling efficient optimization of large and nonlinear networks. This approach demonstrates how careful modeling and control can improve the efficiency, reliability, and cost-effectiveness of heating networks.

Abstract: This paper demonstrates the potential use of affordable, and efficient electrocatalysts, which can maintain the efficiency and stability of platinum-group metals in water-splitting. The study focuses on the optimization and setup of a PEM electrolyzer, alongside the development of new methods for preparing membrane electrode assemblies (MEA) using cost-effective and efficient catalyst materials. The integration of a fibrous membrane layer into the MEA architecture represents a promising design strategy, offering excellent structural and transport properties. Herein, a simple preparation method for modified NiCoP electrocatalysts in the form of carbon fibers is presented, using needleless electrospinning combined with airbrush spraying of an Ir-black solution onto a perfluorosulfonic membrane (Nafion), later pressed together with NiCoP carbon fibers to form a custom-made MEA. For electrochemical testing, custom made MEA was directly evaluated in the PEM electrolyzer setup, providing a preliminary demonstration of overall performance and stability.

Abstract: Infant malnutrition remains a major health problem in West Africa, particularly among children aged 6 months, the crucial period for dietary diversification. Faced with the predominance of imported industrial flours, which are often expensive, local populations are turning to traditional flours. To improve the nutritional quality of these flours, food fortification, recommended by bodies such as the FAO and the WFP, is commonly used. In this study, the mixture design method was used to formulate an affordable complementary flour, enriched in iron, zinc and vitamin C, from under-exploited local plant resources such as Anacardium occidentale kernel fragments and Parkia biglobosa pulp. An augmented centred mixing design with constraints was used to formulate, model and optimise the iron, zinc and vitamin C content of the infant supplement flour. Modelling of the iron content revealed a synergistically interacting cubic model with a desirability of 0.97, and an average iron content of 14.13 mg/100 g. Zinc content was estimated at 5.78 mg/100 g and modelled by a significant quadratic model. The vitamin C content was better represented by a linear model with a synergistic interaction, with a desirability of 0.97 and an average content of 117.6 mg/100 g, well above the standard of 30 mg/100 g. In conclusion, the optimisation has maximised the iron, zinc and vitamin C content of the formulation, offering an improved nutritional solution for combating infant malnutrition.

Abstract: This paper examines the evolution, processes, and optimization of Fused Deposition Modeling (FDM)/Fused Filament Fabrication (FFF) in additive manufacturing, synthesizing insights from existing literature on its mechanical properties and process parameters. Tracing its origins to rapid prototyping in the late 1980s, the paper highlights the advantages of FDM/FFF, such as cost-effectiveness and reduced material waste, while also addressing challenges like limited part strength. It consolidates knowledge on commonly used materials polylactic acid, acrylonitrile butadiene styrene, polycarbonate, and nylon through comparative analyses of their mechanical and thermophysical properties. The review critically assesses key process parameters, including raster angle, layer height, infill density, infill pattern, build orientation, printing speed, and nozzle diameter, drawing from diverse studies to explore their influence on part quality. Key findings include the potential of a 45°/-45° raster angle and a 0.2 mm layer height to enhance tensile strength, as well as the trade-offs associated with higher infill densities, which improve energy absorption but increase printing time. The paper identifies gaps in dimensional accuracy and material innovation, proposing future research directions to advance FDM/FFF applications across industries.