The present study is concerned with the deformation response of an architectured material system, i.e., a 2D-material system created by the topological interlocking assembly of polyhedra. Following the analogy of granular crystals, the internal load transfer is considered along well-defined force networks, and internal equivalent truss structures are used to describe the deformation response. Closed-form relationships for stiffness, strength, and toughness of the topologically interlocked material system are presented. The model is validated relative to direct numerical simulation results. The topologically interlocked material system characteristics are compared with those of monolithic plates. The architectured material system outperforms equivalent size monolithic plates in terms of toughness for nearly all possible ratios of modulus to the strength of the material used to make the building blocks and plate, respectively. In addition, topologically interlocked material systems are shown to provide better strength characteristics than a monolithic system for low strength solids.

Nonlinear interaction of Lamb waves with an imperfect joint of plates for the incidence of the lowest-order symmetric (S0) Lamb wave is investigated by perturbation analysis and time-domain numerical simulation. The imperfect joint is modeled as a nonlinear spring-type interface, which expresses interfacial stresses as functions of the displacement discontinuities. In the perturbation analysis, under the assumption of weak nonlinearity, the second-harmonic generation at the joint is examined in the frequency domain by the thin-plate approximation using extensional waves. As a result, the amplitude of the second-harmonic extensional wave is shown to be in good agreement with the result of the S0 mode in a low-frequency range. However, it is found that the thin-plate approximation does not reproduce the amplification of the second-harmonic S0 mode, which occurs due to the resonance of the joint. Furthermore, the time-domain analysis is performed by the elastodynamic finite integration technique (EFIT). When the amplitude of the incident wave is relatively large, the fundamental wave and the second harmonic exhibit different behavior from the results by the perturbation analysis. Specifically, if the incident amplitude is increased, the peak frequency of the second-harmonic amplitude becomes low. The transient behavior of the nonlinear interaction is also examined and discussed based on the results for the weak nonlinearity.

One-dimensional (1D) modeling is crucial for turbomachinery unsteady performance prediction and system response assessment. The purpose of the paper is to describe a newly developed 1D modeling (turbomachinery dynamic simulator (TURBODYNA)) for turbomachinery. Different from classic 1D modeling, in TURBODYNA, rotor has been meshed and its unsteadiness due to flow field timescale is considered. Instead of direct using of performances maps, source terms are added in Euler equation set to simulate the rotor. By comparing 1D modeling with three-dimensional (3D) computational fluid dynamics (CFD) results, it shows that rotor unsteadiness is indispensable for a better prediction. In addition, different variables response to pulse differently. In the rotor, mass flow is close to quasi-steady while entropy is significantly unsteady. TURBODYNA can capture these features correctly and provide an accurate prediction on pressure wave transportation.

Recent advances in design optimization have significant potential to improve the function of mechanical components and systems. Coupled with additive manufacturing, topology optimization is one category of numerical methods used to produce algorithmically generated optimized designs making a difference in the mechanical design of hardware currently being introduced to the market. Unfortunately, many of these algorithms require extensive manual setup and control, particularly of tuning parameters that control algorithmic function and convergence. This paper introduces a framework based on machine learning approaches to recommend tuning parameters to a user in order to avoid costly trial and error involved in manual tuning. The algorithm reads tuning parameters from a repository of prior, similar problems adjudged using a dissimilarity metric based on problem metadata and refines them for the current problem using a Bayesian optimization approach. The approach is demonstrated for a simple topology optimization problem with the objective of achieving good topology optimization solution quality and then with the additional objective of finding an optimal “trade” between solution quality and required computational time. The goal is to reduce the total number of “wasted” tuning runs that would be required for purely manual tuning. With more development, the framework may ultimately be useful on an enterprise level for analysis and optimization problems—topology optimization is one example but the framework is also applicable to other optimization problems such as shape and sizing and in high-fidelity physics-based analysis models—and enable these types of advanced approaches to be used more efficiently.

Humans as designers have quite versatile problem-solving strategies. Computer agents on the other hand can access large-scale computational resources to solve certain design problems. Hence, if agents can learn from human behavior, a synergetic human-agent problem-solving team can be created. This paper presents an approach to extract human design strategies and implicit rules, purely from historical human data, and use that for design generation. A two-step framework that learns to imitate human design strategies from observation is proposed and implemented. This framework makes use of deep learning constructs to learn to generate designs without any explicit information about objective and performance metrics. The framework is designed to interact with the problem through a visual interface as humans did when solving the problem. It is trained to imitate a set of human designers by observing their design state sequences without inducing problem-specific modeling bias or extra information about the problem. Furthermore, an end-to-end agent is developed that uses this deep learning framework as its core in conjunction with image processing to map pixel-to-design moves as a mechanism to generate designs. Finally, the designs generated by a computational team of these agents are then compared with actual human data for teams solving a truss design problem. Results demonstrate that these agents are able to create feasible and efficient truss designs without guidance, showing that this methodology allows agents to learn effective design strategies.

Machine learning can be used to automate common or time-consuming engineering tasks for which sufficient data already exist. For instance, design repositories can be used to train deep learning algorithms to assess component manufacturability; however, methods to determine the suitability of a design repository for use with machine learning do not exist. We provide an initial investigation toward identifying such a method using “artificial” design repositories to experimentally test the extent to which altering properties of the dataset impacts the assessment precision and generalizability of neural networks trained on the data. For this experiment, we use a 3D convolutional neural network to estimate quantitative manufacturing metrics directly from voxel-based component geometries. Additive manufacturing (AM) is used as a case study because of the recent growth of AM-focused design repositories such as GrabCAD and Thingiverse that are readily accessible online. In this study, we focus only on material extrusion, the dominant consumer AM process, and investigate three AM build metrics: (1) part mass, (2) support material mass, and (3) build time. Additionally, we compare the convolutional neural network accuracy to that of a baseline multiple linear regression model. Our results suggest that training on design repositories with less standardized orientation and position resulted in more accurate trained neural networks and that orientation-dependent metrics were harder to estimate than orientation-independent metrics. Furthermore, the convolutional neural network was more accurate than the baseline linear regression model for all build metrics.

To solve challenging optimization problems with time-consuming objective and constraints, a novel efficient Kriging-based constrained optimization (EKCO) algorithm is proposed in this paper. The EKCO mainly consists of three sampling phases. In Phase I of EKCO, considering the significance of constraints, feasible region is constructed via employing a feasible region sampling (FRS) criterion. The FRS criterion can avoid the local clustering phenomenon of sample points. Therefore, Phase I is also a global sampling process for the objective function in the feasible region. However, the objective function may be higher-order nonlinear than constraints. In Phase II, by maximizing the prediction variance of the surrogate objective, more accurate objective function in the feasible region can be obtained. After global sampling, to accelerate the convergence of EKCO, an objective local sampling criterion is introduced in Phase III. The verification of the EKCO algorithm is examined on 18 problems by several recently published surrogate-based optimization algorithms. The results indicate that the sampling efficiency of EKCO is higher than or comparable to that of the recently published algorithms while maintaining the high accuracy of the optimal solution, and the adaptive ability of the proposed algorithm also be validated. To verify the ability of EKCO to solve practical engineering problems, an optimization design problem of aeronautical structure is presented. The result indicates EKCO can find a better feasible design than the initial design with limited sample points, which demonstrates practicality of EKCO.

High-performance centrifugal compressors presently favor pocket damper seals (PDSs) as a choice of secondary flow control element offering a large effective damping coefficient to mitigate rotor subsynchronous whirl motions. Current and upcoming multiple-phase compression systems in subsea production facilities must demonstrate long-term operation and continuous availability, free of harmful rotor instabilities. Plain annular seals and labyrinth (LABY) seals are notoriously bad choices, whereas a PDS, by stopping the circulation of trapped liquid, operates stably. This paper presents experimental and computational fluid dynamics (CFD) results for the leakage and dynamic force coefficients obtained in a dedicated test facility hosting a fully partitioned PDS (FPPDS), four ribbed and with eight pockets per cavity. The test PDS, operating at a rotor speed 5250 rpm (surface speed 35 m/s) and under a supply pressure/discharge pressure ratio up to 3.2, is supplied with a mixture of air and ISO VG 10 oil whose maximum liquid volume fraction (LVF) is 2.2%, equivalent to a liquid mass fraction of 84%. When supplied with just air (dry condition), the measured leakage increases nonlinearly with supply pressure. Under a wet gas condition, the recorded mass flow increases on account of the large difference in density between the liquid and the gas. CFD-derived mass flow rates for both dry and wet gas conditions agree with the measured ones. The test dry gas PDS produces a direct dynamic stiffness (H R ) increasing with frequency, whereas the direct damping (C) and cross-coupled dynamic stiffness (h R ) coefficients remain relatively constant. The CFD-predicted damping agrees best with the test C albeit overpredicting H R at low excitation frequencies and h R at all frequencies (<175 Hz ∼ twice rotor speed). Under a wet gas condition with LVF = 0.4%, the test force coefficients show great variability over the excitation frequency range; in particular, H R < 0, though growing with frequency due to the large liquid mass fraction. The CFD predictions, on the other hand, produce a dynamic direct stiffness H R > 0 for all frequencies. Both experimental h R and C for the wet gas PDS are larger than their counterparts for the dry gas seal. The CFD-predicted C and h R , wet versus dry, show a modest growth, yet remaining lower than the test data. The CFD-derived flow field for a wet gas condition shows that the seal radial partition walls (ridges) reduce the circumferential flow velocity and liquid accumulation within a pocket. Both the test data and the CFD prediction show that the magnitude of the flexibility function for the PDS test system reduces when the two-component mixture flows through the seal, hence revealing the additional effective damping, more pronounced for the test data rather than that from the predictions. Further work, experimental and CFD based, will continue to advance the technology of wet gas seals while bridging the gap between test data and computational physics model simulations.

Pressure gain combustion (PGC) is considered to be a potential technology to increase the cycle efficiency of gas turbine. As one viable candidate for PGC, rotating detonation engine (RDE) draws more attention due to its significant advances in continuous mode of operation. In practical, one of the basic challenges for RDE application is to reliably initiate detonation wave. For this purpose, both detonation initiation mechanism and enhancement approach are urgently needed to be understood. In this work, a toroidal shock wave focusing detonation initiator is presented. On this basis, the two-dimensional numerical simulations are carried out to investigate the detonation initiation characteristics by using the toroidal shock wave focusing. All of the flame acceleration, shock wave focusing, detonation wave forming, and propagation are analyzed in detail. The numerical results show that the toroidal shock wave focusing initiator developed in this study can rapidly realize the detonation initiation over a short distance and performs significantly better than the traditional smooth or obstructed tube based imitators under different operating conditions. Under the same operating condition, the novel developed initiator decreases time of 59.2% and distance of 84.7% for the smooth tube based initiator, and time of 52% and distance of 78.9% for the obstructed one. Besides, the multifield analysis indicates that both the local explosion induced by shock wave focusing in concave cavity and the entrainment vortex generated by shock wave and jet flame in front of diaphragm are important mechanisms to initiate detonation wave. This study is expected to enhance the understanding of the physical mechanism of shock wave focusing detonation initiation and contribute to the development of detonation propulsion technology.

A CFD-based throughflow solver is applied to the meridional analysis of low-pressure steam turbine modules. The throughflow code inherits its numerical scheme from a state-of-the-art CFD solver (TRAF code) and incorporates real gas capabilities, three-dimensional flow features and spanwise mixing models. Secondary flow effects are introduced via a concentrated vortex model. Tip gap and shroud leakage effects are modelled in terms of source vectors in the system of governing equations. The impact of part-span shrouds and snubbers are accounted for, on a local basis, through suitable body force fields. The AUSM+ -up upwind strategy has been adopted as a basis to construct a numerical flux scheme explicitly suited for throughflow applications. The original formulation has been adapted to handle real gas flows and to embed the treatment of body force fields in a fully consistent framework. The capability of the procedure is assessed by analysing the low-pressure modules of two large steam turbines designed and manufactured by Ansaldo Energia. These 3-stage modules include rotor tip shrouds and part-span snubbers, and feature supersonic flow and large blade twist. Throughflow predictions in terms of main performance figures and radial distributions of flow quantities are compared to experimental data and 3D steady viscous analyses. It will be shown how the proposed throughflow model can be fruitfully used in the early stages of the design as it delivers predictions of comparable accuracy with 3D CFD analyses at a fraction of the computational time.

The design of Ranque–Hilsch vortex tube (RHVT) seems to be interesting for refrigeration and air conditioning purposes in industry. Improving thermal efficiency of the vortex tubes could increase the operability of these innovative facilities for a wider heat and cooling demand to this end; it is of an interest to understand the physical phenomena of thermal and flow patterns inside a vortex tube. In this work, the flow phenomena and the thermal energy transfer in RHVT are studied for three RHVT: straight, divergent, and convergent vortex tubes. A three-dimensional numerical analysis of swirling or vortex flow is performed, verified, and validated against previous experimental and numerical data reported in literature. The flow field and the temperature separation inside an RHVT for different configuration of straight, five angles of divergent hot tube (1 deg, 2 deg, 3 deg, 4 deg, and 6 deg) and five angle of convergent hot tube (0.5 deg, 0.8 deg, 1 deg, 1.5 deg, and 2 deg) are investigated. The thermal performance for all investigated RHVTs configuration is determined and quantitatively assessed via visualizing the stream lines for all three scenarios.

In this paper, we present a computational fluid dynamics (CFD) model to perform single- and two-phase fluid flow simulation on two- and three-dimensional perforated porous media with different perforation geometries. The finite volume method (FVM) has been employed to solve the equations governing the fluid flow through the porous media and obtain the pressure and velocity profiles. The volume of fluid (VOF) method has also been utilized for accurate determination of the volume occupied by each phase. The validity of the model has been achieved via comparing the simulation results with the available experimental data in the literature. The model was used to analyze the effect of perforation geometrical parameters (length and diameter), degree of heterogeneity, and also crushed zone properties (permeability and thickness) on the pressure and velocity profiles. The two-phase fluid flow around the perforation tunnel under the transient flow regime was also investigated by considering a constant mass flow boundary condition at the inlet. The developed model successfully predicted the pressure drop and resultant temperature changes for the system of air–water along clean and gravel-filled perforations under the steady-state conditions. The presented model in this study can be used as an efficient tool to design the most appropriate perforation strategy with respect to the well characteristics and reservoir properties.

The proper thermal diagnostics of the pipeline insulation is an important problem. The heat losses from the pipelines depend distinctly on a quality of this insulation. Changes of weather conditions cause transient accumulation of heat in the pipeline insulation and may cause difficulties during evaluation of the quality of the pipeline thermal insulation. Generally, the goal of this investigation was to identify the heat exchange processes on the external shell of the pipelines and the scale of heat accumulation inside the thermal insulation. This is important because during the calculations of heat losses from the thermal pipelines on the basis of infrared camera measurement results usually a steady thermal state is assumed. In order to determine the distributions of the temperature inside the thermal insulation the calculations of the temperature changes inside the pipeline insulation for real changeable meteorological conditions with use of software Ansys-Fluent and others have been carried out. Both the heat transfer between the inner pipeline tube and outer pipeline shell and heat accumulation inside the pipeline elements were considered. For the pipeline insulation evaluation purpose different coefficients for the analysis of heat accumulation scale were defined and used. The measurement results of the temperature of inner pipeline tube and outer pipeline shell gathered during the operation of the special experimental rig were used as input data for aforementioned numerical simulations. The conclusions resulting from this work may be useful for specialists involved in the examination and technical evaluation of the thermal insulation of pipelines.

The energy, exergy and economic aspects are analyzed of a cycle consisting of a polymer fuel cell, a burner, a reformer and a heat exchanger. Water is used for cooling the fuel cell, and the heated water is used for domestic consumption. The exergy and energy efficiencies of the cycle are calculated and the effects of various cycle parameters on the exergy and energy efficiencies are investigated. To maximize the exergy efficiency while minimizing the cost of electricity generation by the fuel cell, the PSO algorithm is utilized. The results show that increasing the cooling water flow rate has the greatest effect on increasing the energy efficiency of the cycle, while increasing the burner temperature has the greatest effect on increasing the exergy efficiency of the cycle. Moreover, it is shown via multiobjective optimization of the proposed cycle that the exergy efficiency of the cycle increases by 38% and the cost of electricity generation decreases by 18% by applying optimized parameters.

System architecture and modularity decisions are inherent to preliminary concept design. Prior modularity research has considered minimizing interactions between modules and increasing the commonality among modular product variants. Effective approaches include function structure partitioning guidelines, affinity analysis or matrix clustering algorithms. We consider here designs with field constraints, such as situations when elements cannot be placed in certain regions such as a high temperature field, a high pressure field, a high magnetic field, etc. which place constraints on modularity choices. Practical design guidelines are developed here for modularity considering field constraints. Two types of guidelines are proposed, field separation and concept generation. The field separation guidelines propose zonal boundaries within which a system modules need be confined. The concept generation guidelines propose how to violate the field constraints through new concepts. Moving functionality from one side of a field boundary to the other is nontrivial and involves new concept generation for the modules to function at the higher or lower field values. The guidelines are defined and illustrated via multiple common examples as well as two extended case studies. We demonstrate the approach using field boundaries on an electric motor controller and on a medical contrast injector, and also use of fields to generated novel concepts. The guidelines support for modularity concept and embodiment decisions.

Liquefied Petroleum Gas (LPG) is commonly used in domestic kitchen due to its low health and environmental impacts and high heat content compared to other traditional fuels. Growing demand of LPG, notwithstanding its limited reserve, influences the need for performance improvement of the LPG cook stoves. In a LPG cook stove, the fuel-air mixture is prepared in a self-aspirated burner and burns as a rich premixed flame at the burner head. The fuel-air ratio of the reactant mixture influences the burning characteristics and thermal efficiency of the cook stove. This part of work deals with a numerical study of flow and mixing characteristics in the mixing tube and burner assembly of a domestic LPG cook stove. The fuel is injected axially through a narrow nozzle inside the mixing tube causing entrainment of ambient air through the side-ports and end-port. The effects of different side-port geometry, nozzle position, inlet fuel pressure and nozzle throat diameter on the air ingress have been systematically investigated and the optimum design has been identified. Results of the study provide important information on the design of the practical nozzle and mixing tube assembly of high-efficiency LPG stoves.

The simultaneous control of force and motion is important in everyday activities when humans interact with objects. While many studies have analyzed the control of movement within a perturbing force field, few have investigated its dual aspects of controlling a contact force in nonisometric conditions. The mechanism by which the central nervous system controls forces during movements is still unclear, and it can be elucidated by estimating the mechanical properties of the arm during tasks with concurrent motion and contact force goals. We investigate how arm mechanics change when a force control task is accomplished during low-frequency positional perturbations of the arm. Contrary to many force regulation algorithms implemented in robotics, where contact impedance is decreased to reduce force fluctuations in response to position disturbances, we observed a steady increase of arm endpoint stiffness as the task progressed. Based on this evidence, we propose a theoretical framework suggesting that an internal model of the perturbing trajectory is formed. We observed that force regulation in the presence of predictable positional disturbances is implemented using a position control strategy together with the modulation of the endpoint stiffness magnitude, where the direction of the endpoint stiffness ellipse's major axis is oriented toward the desired force.

A new reduction method is proposed to investigate the behavior stability of rotor-bearing systems subject to a multifrequency rotational motion of their base. Combining the modal analysis and the construction of specific dynamic Ritz vectors, this method is able to deal with complex rotordynamics characteristics such as nonproportional damping, nonself-adjoint matrices, or time-varying parametric coefficients. This paper focuses first on assessing the accuracy and efficiency of the reduction method by computing time history and spectral responses of full and reduced models due to multifrequency base excitations. Its main potential is then highlighted in the parametric stability analysis through Floquet theory. The proposed numerical examples are composed with academic and industrial rotors, both modeled with one-dimensional Timoshenko beam finite element and supported by hydrodynamic journal bearings.

Due to imperfection in the manufacturing process and in-service wear, fan blades in a turbofan engine do not have the same geometry. This lack of symmetry inevitably leads to difficulties in predicting the fan blades’running geometry in an assembly as blade variability is amplified by the aerodynamic and centrifugal loading. This variability can lead to an aeromechanical phenomenon termed ‘Alternate Passage Divergence’ (APD). As the name suggests, it manifests itself as alternating geometric and aerodynamic patterns in a fan assembly during operation. APD can potentially influence fan performance, stability, and Multiple Pure Tone (MPT)/’Buzz Saw’ noise and is therefore an important area of research. For this study, the APD phenomenon is purposely triggered on a transonic fan blade. A full assembly computational domain with one mis-staggered blade is used to examine whole assembly performance and untwist behaviour with APD. In particular, the influence of fan blade stiffness on APD behaviour is examined. The behaviour of the current blade is compared to that of a blade used in a precursor study and it is found that, under certain conditions, the blades show similar behaviour even though they have distinctly different geometry features. This illustrates that it is important to understand the phenomenon as accurate prediction of running geometry is vital at early design stage.

Non-engine order rotor blade vibration is an aeroelastic phenomenon of major interest for compressor designers resulting from an excitation of rotor blade modes through aerodynamic instabilities. Indicators for a comparable type of instability, caused by propagating acoustic modes have been observed in an experimental multistage high-speed compressor by Safran Helicopter Engines. It is intended to understand the cause for these instabilities by combining experimental data and numerical simulations. Unsteady pressure measurements were carried out by case-mounted and stator-mounted transducers. Rotor tip timing and magnet-coil sensor systems were installed to measure the blade vibrations. Experimental results show non-engine order signatures in the unsteady pressure signal coherent to the shifted frequency of blade vibrations. In the present paper the waveform of these oscillations is analysed in detail, showing a dominant propagating acoustic mode interacting with vibrations of Rotor-2. The root cause for the non-synchronous oscillations is identified as an acoustic mode that is cut-off downstream of Rotor-3. During the test, the mode changes its frequency and circumferential order, affecting the amplitude of associated blade vibrations.