Current Design Methodologies Research Articles

Evolutionary Perspectives on Neural Network Generations: A Critical Examination of Models and Design Strategies

Abstract: In the last few years, Neural Networks have become more common in different areas due to their ability to learn intricate patterns and provide precise predictions. Nonetheless, creating an efficient neural network model is a difficult task that demands careful thought of multiple factors, such as architecture, optimization method, and regularization technique. This paper aims to comprehensively overview the state-of-the-art artificial neural network (ANN) generation and highlight key challenges and opportunities in machine learning applications. It provides a critical analysis of current neural network model design methodologies, focusing on the strengths and weaknesses of different approaches. Also, it explores the use of different deep neural networks (DNN) in image recognition, natural language processing, and time series analysis. In addition, the text explores the advantages of selecting optimal values for various components of an Artificial Neural Network (ANN). These components include the number of input/output layers, the number of hidden layers, the type of activation function used, the number of epochs, and the model type selection. Setting these components to their ideal values can help enhance the model's overall performance and generalization. Furthermore, it identifies some common pitfalls and limitations of existing design methodologies, such as overfitting, lack of interpretability, and computational complexity. Finally, it proposes some directions for future research, such as developing more efficient and interpretable neural network architectures, improving the scalability of training algorithms, and exploring the potential of new paradigms, such as Spiking Neural Networks, quantum neural networks, and neuromorphic computing.

Sustainable space technologies—Strategies toward a predictive aerothermal design of re-useable space transportation systems

This paper presents a review of current aerothermal design and analysis methodologies for spacecraft. It briefly introduces the most important system architectures, including rockets, gliders, and capsule-based configurations, and gives an overview of the specific aerothermal and thermo-chemical effects that are encountered during their different flight phases and trajectories. Numerical and experimental design tools of different fidelity levels are reviewed and discussed, with a specific focus placed on the present limitations and uncertainty sources of models for the wide range of physical phenomena that are encountered in the analyses. This includes high temperature thermodynamics, chemical effects, turbulence, radiation, and gasdynamic effects. This is followed by a summary of current predictive capabilities and research foci, with missing capabilities identified. Finally, a future strategy toward an efficient and predictive aerothermal design of re-useable space transportation systems is proposed.

Exploring Embedded System Tackling Design Challenges Methodology and Optimization Strategies

The current design methodologies are no longer relevant due to the complex nature of embedded products and the need for quick development. The relevance of the present state of embedded system technology is highlighted, emphasizing the need to resolve design problems, implement optimization techniques, and create development processes for embedded systems. We fully focus on the safety and security of your design when we are solving a problem, especially for embedded systems applications that are critical to mission success. A strong understanding of embedded system hardware and software design is necessary in order to implement the strategies and methodologies of embedded system development. This is due to the fact that embedded systems comprise both software and hardware. These kinds of constraints include things like time constraints that are applied in real time, interactive designs, validation, and extensive testing.

Machine learning prediction model for the axial strength of longitudinal branch plate-to-CHS T-connections

Circular hollow sections (CHS) are widely used in construction, particularly in space structures. This paper aims at investigating the feasibility of employing a machine leaning approach to predict the axial strength of longitudinal branch plate-to-CHS connections subjected to branch loadings. Using nonlinear finite element analysis, numerical models are developed for the connections subjected to tensile or compressive branch loading. The difference between the load–displacement responses for the connections under compressive and under tensile loading is established. Furthermore, the influence of various geometric and material parameters on the strength is investigated in depth. Using the results, two different strength models for the prediction of both tensile and compressive strength of longitudinal branch plate-to-CHS T-connections are developed. One prediction model is attained using traditional nonlinear regression analysis, and the other model is attained through symbolic regression. Finally, theoretical predictions are compared with numerical results, and strengths predicted with current design methodologies.

Analytical method for evaluating the tensioned membrane effect in the geosynthetic-reinforced unsaturated embankment soil subjected to localized sinkholes

In karst regions, the occurrence of sinkholes is a common cause of roadway failures, leading to significant embankment deformation and instability. The advantages of geosynthetic reinforcement have been proven in practice in numerous countermeasures to lessen the impact of localized sinkholes, and associated design methodologies have been developed. However, some existing design methods assume there is no deformation in geosynthetics embedded in soils, and the effect of matric suction in soils is not considered either. This study suggested an analytical approach that took into account the strain in the geosynthetic reinforcements as well as the impact of matric suction. The developed design method was used to conduct parameter studies under various matric suction conditions. The method was validated against the experimental results, and the feasibility of the developed method was also compared with the existing design methods. The findings demonstrate that, in comparison to the method created here, existing design methodologies produce more conservative results. The strain and the deflection of the geosynthetic calculated using the suggested method are lower than those provided by the current design methodologies.

Implementation of OTA in 90nm Technology with Bandgap Reference Application

Operational Transconductance Amplifier (OTA) is the basic building block of Analog circuit with linear input/output characteristics. Because it associates closely linked parameters like noise and settling performance, the current design methodologies for two-stage OTAs frequently result in sub-optimal solutions. The study discusses the implementation of an Operational transconductance amplifier (OTA) with a bandgap reference (BGR) circuit and its importance in providing a stable and accurate reference voltage for various analog and digital circuits. The combination of OTA and BGR provides a stable and accurate reference voltage that can be used in various analog and digital circuits. OTAs are used to drive high capacitive loads. To scale the dimensions of Complementary Metal Oxide Semiconductor (CMOS) technology requires proportional scaling, also taking supply voltage into consideration. In addition to the simplicity and strong performance that the Band Gap Reference (BGR) achieves as a load for the OTA circuit, the architecture of OTA with BGR as load is easily adaptable to attain high accuracy. Because the performance of the used amplifier has a significant impact on the OTA’s performance, consideration must be given to the design and its architecture. Therefore, it is essential to carry out a comparison analysis in order to get the optimal results of the outputs of OTA having BGR as the load.

Design approach and its optimization for bolted steel plate strengthened RC beams considering the effect of connection bolt slip

The use of bolted steel plates (BSP) has been found to be an effective method for increasing the flexural capacity of reinforced concrete (RC) beams. However, slip in bolt connections is an inevitable occurrence that affects the mechanical behavior of the BSP beam. While prior studies have thoroughly examined the effects of bolt connection slip on the flexural capacity of BSP beams, assuming unlimited slip, experimental results revealed that bolt connection failure takes place when slip surpasses a specific threshold. The current design methodology does not accommodate this finding. To address this issue, a simplified design method for practical applications is proposed. Comprehensive parameter studies were conducted on the enhanced moment capacity ratio and maximum bolt slip, involving a total of 13,824 and 276,480 calculations, respectively. It is found that several parameters (e.g., the aspect ratio of RC beam, the tensile longitudinal reinforcement ratio, the height of steel plate) affect the two above-mentioned indexes. Equations were then developed to determine the enhanced moment capacity ratio and maximum slip. It is noteworthy that the recommended effective yield slip (0.55 mm) is obtained based on adhesive anchor, which should be checked when other types of anchors would be used. Relying on these equations, an optimized design approach for BSP beams is introduced that avoids complex iterative processes. The effectiveness of the optimized design methodology is substantiated through a practical example.

Strategies employed in the design and optimization of pump as turbine runner

The low hydraulic efficiency of the pump as turbine (PAT) poses a major challenge to the runner design. This paper introduces a new strategy for designing and optimizing the performance of PAT runner. The strategy process is mainly divided into four steps, namely blade parametric design, significance evaluation of parameters, construction of optimizing surrogate model and optimization of improved PSO algorithm. Compared with the current design methodologies, not only does this optimization strategy propose the system design and optimization process, but also it breaks down the process into well-defined steps and simplifies them. The hydraulic efficiency of PAT optimized by this optimization strategy reaches 84.76%, which improves 3.93%. A test bench has been built to test the optimizing strategy. The maximum difference of experimental data is 1.5%, which meets the requirements of practical engineering application, and verifies the reliability of the optimization strategy process. The optimizing strategy process can be extended to apply to pump, fan, conventional turbine and other blade fluid models, and it can be further extended to apply to the optimization design of 3D blades with a high specific speed.

Impact of concentration polarization on membrane gas separation processes: From 1D modelling to detailed CFD simulations

Membranes are one key technology for gas separation. Current design methodologies make use of a 1D set of mass balance equations with membrane taken as the only mass transfer resistance (no concentration polarization hypothesis). This strategy has however to be reconsidered with emerging high-performance membrane materials. Accordingly, a rigorous CFD simulation approach is developed in order to correctly compute their separation efficiency. The results reveal that, compared to CFD, in the presence of concentration polarization, 1D models largely overestimate the separation performance of high-performing membranes. Based on a generic modelling of mass transfer in the gas phase under concentration polarization, a novel 1D modified approach is reported, showing similar predictions compared to CFD simulations, but offering very fast computations. This strategy is of interest for the simple and systematic design of membrane gas separation processes, including process synthesis approaches.

Architectural Glass under Climatic Actions and Fire: Review of State of the Art, Open Problems and Future Perspectives

A critical issue in the design of structural glass elements in buildings is represented by the evaluation of thermally induced stresses and strains. For both climatic actions and fire, thermal stresses represent one of the main causes of premature failure, due to the high sensitivity of glass to temperature gradients. Thermal loads pose a severe safety risk for glass, due to their uneven distribution but also the lack of knowledge on the modification of mechanical properties with temperature. In design practice, approximate tools are used to describe temperature fields in glazing, which do not adequately estimate the thermally induced stresses. Additionally, the existing standards prescribe different methods for the calculation of both the temperature field and the consequent stress, usually based on strong simplifying assumptions, and there is a lack of uniformly defined procedures. Here, an accurate review of the state of the art on glass elements exposed to thermal actions, from both the scientific and the regulatory perspectives, is presented. Reference is made first to the evaluation of the thermal actions, and further to the proper assessment of both the temperature distribution and the consequent thermal stress. The paper also emphasizes open problems and future perspectives related to these topics, to evidence areas of research that should be strengthened and possible future enhancements to the current design and assessment methodologies, which should also be introduced in a regulatory framework.

Improved analysis and design for surcharging

Surcharging or preloading is one of the oldest and most widely used in situ ground improvement techniques to strengthen weak compressible clays and to remove undesirable settlements prior to erection of permanent infrastructure. Recently, the ground improvement technique has been adopted in combination with many other techniques. The current design methodology for surcharging based on the concept of average degree of consolidation is inadequate, as all the primary consolidation settlement to be induced by the permanent load is not eliminated by the surcharging process. The problem is quantitatively analysed using Terzaghi's one-dimensional consolidation theory. The proportion of residual primary consolidation settlement is quantified, and the starting time of the re-occurrence of primary consolidation on exertion of the permanent load is determined. The effect of a change in surcharging duration on the proportion of residual settlement is quantified. The results are presented in design charts to facilitate application of the results by practicing geotechnical engineers.

An analytical review of current S-box design methodologies, performance evaluation criteria, and major challenges

An analytical review of current S-box design methodologies, performance evaluation criteria, and major challenges

Estimates of Internal Forces in Torsionally Braced Steel I-Girder Bridges Using Deep Neural Networks

The bracing components in steel I-girder bridge systems are essential structural components for the bridges to restrain their rotation due to lateral torsional buckling (LTB). Current design specifications require bracing components to be installed to prevent I-girder sections from unexpectedly twisting due to instability. To estimate the bracing internal forces acting on the bracing elements, we can use approximate design equations that provide considerably conservative design values. Otherwise, it is necessary to conduct a thorough finite element analysis considering initial imperfections to obtain accurate bracing internal forces in the steel I-girder bracing systems. This study aims to provide estimation models based on deep neural network (DNN) algorithms to more accurately estimate the internal forces acting on the bracing element compared with the current design methodology when LTB occurs. This is conducted by constructing structural response data based on the geometrically nonlinear analysis with imperfections to provide accurate bracing internal forces, namely bracing moments (Mbr) and bracing forces (Fbr). To propose prediction models, 16 input and three output variables were selected for training the structural response data. Furthermore, a parametric study on the hyperparameters used in DNN models was analyzed for the number of hidden layers, neurons, and epochs. Based on statistical performance indices (i.e., RMSE, MSE, MAE, and R2), the estimated values using DNN models were evaluated to determine the best prediction models. Finally, DNN models that more accurately estimate internal forces (Mbr, Fbr) in bracing elements, and that provide the best prediction results depending on hyperparameters (numbers of hidden layers, neurons, and epochs), are proposed.

Methodology for the Redesign of Compressor Blades Undergoing Nonlinear Structural Interactions: Application to Blade-Tip/Casing Contacts

Abstract Over the past decade, the drive towards more efficient aircraft engines has pushed the boundaries of operating ranges far beyond a linear structural context. Nonlinear interfaces, such as blade-tip/casing contacts, are to be expected in nominal operating conditions. However, current blade design methodologies still rely on empirical structural considerations, often linear, which may lead to costly redesign operations. This work aims at proposing a methodology for the redesign of blades undergoing nonlinear structural interactions. A three-step redesign process is considered: (1) parameterization of an existing blade, (2) update of blade parameters with respect to a surrogate performance criterion, and (3) performance check of the optimized blade. An original two-way parameterization method is proposed to parameterize existing blades and generate models from blade parameters. As a proof-of-concept, the redesign of the NASA compressor blade rotor 37 and fan blade rotor 67 with respect to blade-tip/casing contacts is considered. High-fidelity parameterized models of the initial blades are obtained and their dynamic response to contact interactions are analyzed. Geometries are updated with respect to their clearance consumption, as its minimization has shown beneficial effects on the considered contact interactions. The proposed methodology allows us to better assess the relevance of this performance criterion in the context of blade-tip/casing contacts.

An enhanced master–slave control for accurate load sharing among parallel standalone AC microgrids

SummaryWith the advancement of power electronic converters over the last few decades, the capability to integrate solar photovoltaics, wind, diesel generators, batteries, ultracapacitor, and electric vehicles with standalone networks or utility grids has increased. For the reliable operation of standalone networks, accurate load sharing among these sources is required. As a result, several load sharing schemes with inherent merits and related demerits are presented in the literature. This work intends to address shortcomings of earlier reported control schemes with the inheritance of associated merits using the proposed control scheme. The proposed enhanced Master–Slave control, with advanced droop control scheme, is illustrated for Master voltage source inverter with resistive line impedance and improves the steady‐state response by improving voltage and frequency profile. It reduces the voltage drop caused by load effect and droop effect, allowing the voltage and frequency regulation during load changes and distinct load conditions such as balanced linear load, unbalanced linear load, and balanced nonlinear load in standalone alternating current (AC) microgrids. Furthermore, Slave voltage source inverters independently compensate for local load demand. This paper presents precise current control and voltage control design methodology and a mathematical model for voltage control and their performance analysis in time and frequency domain. Finally, the proposed enhanced Master–Slave control for accurate load sharing among parallel standalone AC microgrids validated using real‐time simulator and experimental results is provided to verify the effectiveness of the proposed control scheme.

Two-Step Robust Design of LLC Converter With Corner Judgment and Greedy Algorithm

Negative effects caused by parameter fluctuation (PF) of resonant elements have not been thoroughly considered in the current design methodologies of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LLC</i> converters. In this letter, typical negative PF effects are revealed as restriction violation and optimization degradation. Then, a novel two-step robust design methodology is proposed for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LLC</i> converters. The first step is to effectively establish more reliable solving domain considering PF effects on all the restrictions by the proposed corner judgment method. In the second step, the lowest fluctuation efficiency as the novel concept is proposed and taken as the global optimization goal of efficiency due to optimization degradation, with a greedy algorithm proposed for simplicity and effectiveness. Simulation and experimental results have proved the validity and superiority of the proposed design methodology.

Environmental contours for wind-resistant bridge design in complex terrain

Accurate estimation of the extreme wind fields is crucial for long-span bridge design. The current practice is focused on estimating the extreme mean wind speed, neglecting the inherent uncertainty in the turbulence model parameters. However, full-scale measurements on bridges show that such uncertainties are significant and should be considered in design. Here, the environmental contour method (ECM) is used to obtain long-term extreme wind fields considering uncertainties from the mean wind speed, turbulence intensities and spectral parameters measured at the Sulafjord Bridge site. Design contours of combinations of wind field parameters are obtained for target return periods of 4, 50 and 100 years. The contours are based on a proposed probabilistic modeling strategy that combines hindcast mesoscale simulations and field measurements. The contour estimates are also compared with state-of-the-art design values from the design recommendations. It is concluded that the environmental contours provide a more complete and yet intuitive description of the wind field at the bridge site compared to the current design methodology. The ECM is found suitable for obtaining design wind fields at new long-span bridge sites as it makes use of the limited site data more efficiently and it is still easy-to-use for the practicing engineer.

Codesign of Architecture, Control, and Scheduling of Modular Cyber-Physical Production Systems for Design Space Exploration

Design space exploration (DSE) of cyber-physical production systems (CPPS) is a search problem in the space of potential compositional configurations. Current design methodologies follow the separated design paradigm in which the architecture, control, and scheduling are separately designed. Optimization of each part considers only the corresponding goals of interest and overlooks other aspects by adopting gross assumptions, which makes it difficult to determine the global optimal solution for a given system. To address this problem, in this article, we propose a codesign method that considers the design spaces of architecture, control, and scheduling as monolithic, mixed discrete-continuous spaces. We formulate DSE as an optimization problem and propose a generic iterative algorithm schema involving simulation in the loop to solve the abovementioned new problem. To illustrate the effectiveness of the proposed method, a real single-stage reducer assembly production system is considered. The design results demonstrate that our method provides better solutions than does the separated design method.

Application of Probabilistic principles to Set-Based Design for the optimisation of a hybrid-electric propulsion system

Current research in hybrid-electric aircraft propulsion has outlined the increased complexity in design when compared with traditional propulsion. However, current design methodologies rely on aircraft-level analysis and do not include the consideration of the impact of new technologies and their uncertainty. This can be a key factor for the development of future hybrid-electric propulsion systems. In this paper, we present a methodology for exploring the design space using the principles of Set-Based Design, which incorporates probabilistic assessment of requirements and multidisciplinary optimisation with uncertainty. The framework can explore every design parameter combination using a provided performance model of the system under design and evaluate the probability of satisfying a minimum required figure of merit. This process allows to quickly discard configurations incapable of meeting the goals of the optimiser. A multidisciplinary optimiser then is used to obtain the best points in each surviving configuration, together with their uncertainty. This information is used to discard undesirable configurations and build a set of Pareto optimal solutions. We demonstrate an early implementation of the framework for the design of a parallel hybrid-electric propulsion system for a regional aircraft of 50 seats. We achieve a considerable reduction to the required function evaluations and optimisation run time by avoiding the ineffective areas of the design space but at the same time maintaining the optimality potential of the selected sets of design solutions.

Development of a Gaussian Process Model as a Surrogate to Study Load Bridging Performance in Racked Pallets

Current pallet design methodology frequently underestimates the load capacity of the pallet by assuming the payload is uniformly distributed and flexible. By considering the effect of payload characteristics and their interactions during pallet design, the structure of pallets can be optimized and raw material consumption reduced. The objective of this study was to develop a full description of how such payload characteristics affect load bridging on unit loads of stacked corrugated boxes on warehouse racking support. To achieve this goal, the authors expanded on a previously developed finite element model of a simplified unit load segment and conducted a study to screen for the significant factors and interactions. Subsequently, a Gaussian process (GP) regression model was developed to efficiently and accurately replicate the simulation model. Using this GP model, a quantification of the effects and interactions of all the identified significant factors was provided. With this information, packaging designers and researchers can engineer unit loads that consider the effect of the relevant design variables and their impact on pallet performance. Such a model has not been previously developed and can potentially reduce packaging materials’ costs.