Network-based Design Research Articles

Designing Connectivity-Guaranteed Porous Metamaterial Units Using Generative Graph Neural Networks

Abstract Designing 3D porous metamaterial units while ensuring complete connectivity of both solid and pore phases presents a significant challenge. This complete connectivity is crucial for manufacturability and structure-fluid interaction applications (e.g., fluid-filled lattices). In this study, we propose a generative graph neural network-based framework for designing the porous metamaterial units with the constraint of complete connectivity. First, we propose a graph-based metamaterial unit generation approach to generate porous metamaterial samples with complete connectivity in both solid and pore phases. Second, we establish and evaluate three distinct variational graph autoencoder (VGAE)-based generative models to assess their effectiveness in generating an accurate latent space representation of metamaterial structures. By choosing the model with the highest reconstruction accuracy, the property-driven design search is conducted to obtain novel metamaterial unit designs with the targeted properties. A case study on designing liquid-filled metamaterials for thermal conductivity properties is carried out. The effectiveness of the proposed graph neural network-based design framework is evaluated by comparing the performances of the obtained designs with those of known designs in the metamaterial database. Merits and shortcomings of the proposed framework are also discussed.

Product Design Incorporating Competition Relations: A Network-Based Design Framework Considering Local Dependencies

Abstract System design has been facing the challenges of incorporating complex dependencies between individual entities into design formulations. For example, while the decision-based design framework successfully integrated customer preference modeling into optimal design, the problem was formulated from a single entity’s perspective, and the competition between multiple enterprises was not considered in the formulation. Network science has offered several solutions for studying interdependencies in various system contexts. However, efforts have primarily focused on analysis (i.e., the forward problem). The inverse problem still remains: How can we achieve the desired system-level performance by promoting the formation of targeted relations among local entities? In this study, we answer this question by developing a network-based design framework. This framework uses network representations to characterize and capture dependencies and relations between individual entities in complex systems and integrate these representations into design formulations to find optimal decisions for the desired performance of a system. To demonstrate its utility, we applied this framework to the design for market systems with a case study on vacuum cleaners. The objective is to increase the sales of a vacuum cleaner or its market share by optimizing its design attributes, such as suction power and weight, with the consideration of market competition relations, such as inter-brand triadic competition involving three products from different brands. We solve this problem by integrating an exponential random graph model (ERGM) with a genetic algorithm. The results indicate that the new designs, which consider market competition, can effectively increase the purchase frequency of specific vacuum cleaner models and the proposed network-based design method outperforms traditional design optimization.

A design method based on neural network to predict thermoacoustic Stirling engine parameters: Experimental and theoretical assessment

A neural network-based design method for thermoacoustic Stirling engines (TASE) is presented in this work. The main goal of the article is to predict the design parameters (Length of resonator (stub) (LR), Length of inertance tube (Lin), Mean pressure (Pm)) of the thermoacoustic Stirling engine in such a way that the dynamic instability is guaranteed. In this regard, the governing equations of the engine are presented first. Next, with the help of the data extracted from the simulation of the governing equations of the engine, a neural network with the structure of 9–35-3 is trained. The effectiveness of the artificial neural network (ANN) structure is explored via regression and MSE (mean square error) analyses. Next, in order to evaluate and validate the neural network, the experimental data of the constructed thermoacoustic Stirling engine (SUTech-SR-4) is considered. It is important to note that the SUTech-SR-4 is introduced for the first time in this work. Investigations have shown that the neural network could estimate the engine design parameters well and with an acceptable approximation in such a way that the dynamic instability of the engine is also established. Besides, the value of the estimated design parameters for the engine made by the neural network was close to the experimental values. Following, the developed engine is introduced and its performance is discussed. The developed engine has been able to produce a power equivalent to 0.346 W at a pressure of 1 bar (with air working fluid) and a working frequency of 15.9 Hz. It should be noted that the method presented in this article can be used for other types of thermoacoustic Stirling engines. Moreover, the selection of design parameters for the presented method will depend on the physics of the engine and the available facilities.

Non-symmetric plate-lattices: Recurrent neural network-based design of optimal metamaterials

The elastic response of plate-lattices with cubic symmetry can reach the upper, isotropic Hashin-Shtrikman bound. Here, our primary objective is the design of stiff lattices with tailored properties beyond any symmetry or isotropy. We propose a general construction method of plate-lattices, defined as a sequence of plates. We present the large elastic property space, accessible via the geometric control offered by our construction method. Building upon this general formulation, we provide a large-scale comparison to the property spaces of truss- and shell-lattices. Additionally, we observe a distinct correlation between plate orientation and stiffness. To tailor the properties of plate-lattice, we employ a recurrent neural network mapping from a given property to the corresponding structure. Our results highlight the effectiveness of this inverse model in creating plate-lattices with desired properties. We believe that this work holds a paradigm shift in field of inverse design by enabling the efficient customization of the stiffest family of metamaterials.

Neural Network-Based Design of Two-Focal-Spot Terahertz Diffractive Optical Elements

Neural Network-Based Design of Two-Focal-Spot Terahertz Diffractive Optical Elements

Graph Neural Network-Based Design Decision Support for Shared Mobility Systems

Abstract Emerging shared mobility systems are gaining popularity due to their significant economic and environmental benefits. In this paper, we present a network-based approach for predicting travel demand between stations (e.g., whether two stations have sufficient trips to form a strong connection) in shared mobility systems to support system design decisions. In particular, we answer the research question of whether local network information (e.g., the network neighboring station’s features of a station and its surrounding points of interest (POI), such as banks, schools, etc.) would influence the formation of a strong connection or not. If so, to what extent do such factors play a role? To answer this question, we propose using graph neural networks (GNNs), in which the concept of network embedding can capture and quantify the effect of local network structures. We compare the results with a regular artificial neural network (ANN) model that is agnostic to neighborhood information. This study is demonstrated using a real-world bike sharing system, the Divvy Bike in Chicago. We observe that the GNN prediction gains up to 8% higher performance than the ANN model. Our findings show that local network information is vital in the structure of a sharing mobility network, and the results generalize even when the network structure and density change significantly. With the GNN model, we show how it supports two crucial design decisions in bike sharing systems, i.e., where new stations should be added and how much capacity a station should have.

Designing Theoretical Shipborne ADCP Survey Trajectories for High-Frequency Radar Based on a Machine Learning Neural Network

A machine learning neural network-based design for shipborne ADCP navigation is proposed to improve the quality of high-frequency radar measurements. In traditional inversion algorithms for HF radars, sea surface velocity is directly extracted from electromagnetic echoes without constraints from oceanographic processes. Hence, we incorporated oceanographic information from observational data into seabed radar inversion results via an LSTM neural network model to enhance data accuracy. Through a series of numerical simulation experiments, we showed improved data accuracy and feasibility by incorporating both fixed-point and navigation observational data. The results indicate a significant reduction in (related) errors. This study has implications for guiding future navigation observations.

Automatic Gripper-Finger Design, Production, and Application: Toward Fast and Cost-Effective Small-Batch Production

Tactile robot-based assembly line adaptation of new products is still limited by the manual redesign, manufacturing, and exchange of the end-effector setup since the gripper fingers must often be adapted to the geometry of the product components to ensure a successful assembly process. In this article we present an automatic finger design, production, and evaluation pipeline, developed to improve this adaptation process. Two different form-closure–based design principles have been implemented to automatically generate the fingertip geometry: a projected surface representation-based approach as well as a Bézier surface fitting strategy. The resulting fingertips are printed via an automatic production unit and experimentally evaluated based on pick and insertion tasks for three different manipulation objects. To illustrate the potential usage of the introduced design methods for machine learning-based fingertip design approaches, we set up the training and testing process for a neural network-based design method. The proposed automatic fingertip design, production, and application framework represent a step further toward small-batch–size production, since the assembly adaptation effort, flexibility, and scalability are significantly improved.

A Problem-tailored Adversarial Deep Neural Network-Based Attack Model for Feed-Forward Physical Unclonable Functions

With the exceeding advancement in technology, the sophistication of attacks is considerably increasing. Standard security methods fall short of achieving the security essentials of IoT against physical attacks due to the nature of IoTs being resource-constrained elements. Physical Unclonable Functions (PUFs) have been successfully employed as a lightweight memoryless solution to secure IoT devices. PUF is a device that exploits the integrated circuits’ inherent randomness originated during the fabrication process to give each physical entity a unique identifier. Nevertheless, because PUFs are vulnerable to mathematical clonability, Feed-Forward Arbiter PUF (FF PUF) was introduced to withstand potential attack methods. Motivated by the necessity to expose a critical vulnerability of the standard FF PUFs design, we introduce a problem-tailored adversarial model to attack FF PUF design using a carefully engineered loop-specific neural network-based design calibrated and trained using FPGA-based in-silicon implementation data to exhibit real-world attacking scenarios posed on FF PUFs, in addition to applying simulated data. The empirical results show that the proposed adversarial model adds outperforming results to the existing studies in attacking FF PUFs, manifesting the improved efficiency in breaking FF PUFs. We demonstrate our high-performing results in numerical experiments of language modeling using the deep Neural Networks method.

GANDSE: Generative Adversarial Network-based Design Space Exploration for Neural Network Accelerator Design

With the popularity of deep learning, the hardware implementation platform of deep learning has received increasing interest. Unlike the general purpose devices, e.g., CPU or GPU, where the deep learning algorithms are executed at the software level, neural network hardware accelerators directly execute the algorithms to achieve higher energy efficiency and performance improvements. However, as the deep learning algorithms evolve frequently, the engineering effort and cost of designing the hardware accelerators are greatly increased. To improve the design quality while saving the cost, design automation for neural network accelerators was proposed, where design space exploration algorithms are used to automatically search the optimized accelerator design within a design space. Nevertheless, the increasing complexity of the neural network accelerators brings the increasing dimensions to the design space. As a result, the previous design space exploration algorithms are no longer effective enough to find an optimized design. In this work, we propose a neural network accelerator design automation framework named GANDSE, where we rethink the problem of design space exploration, and propose a novel approach based on the generative adversarial network (GAN) to support an optimized exploration for high-dimension large design space. The experiments show that GANDSE is able to find the more optimized designs in negligible time compared with approaches including multilayer perceptron and deep reinforcement learning.

Resonant Network-Based MVDC Circuit Breaker Topologies With Fast Current Breaking Capability

This paper presents two improved circuit breaker (CB) topologies based on resonant networks for medium voltage DC (MVDC) applications, such as shipboard power distribution and subsea oil and gas production. The topologies feature simple LC resonant networks that efficiently suppress fault current in just a few tens of microseconds, making them suitable for integration with existing solid-state or hybrid DC CB. Moreover, they also help to minimize the CB energy and reduce (if not eliminate) the requirement for protection devices like surge arresters during fault events. The proposed topologies are first tested using Typhoon Hardware-in-the-Loop (HiL) simulations at 1.2 MW (6 kV, 200 A) power levels and thyristor-based solid-state circuit breakers (SSCB). Then, scaled-down laboratory prototypes, also based on thyristor-based SSCB, are experimentally validated. The results show that both topologies quickly bring the current of the CB’s main switch to 0 A. The main circuit pathway requires only a small inductor and switch, resulting in a steady-state efficiency of >99.9%. The resonant network-based design allows for a fast response time of less than 50 μs using a simple and low-cost auxiliary circuit, eliminating the need for complex charging circuitry and protection devices typically used to suppress voltage surges during faults.

GAN-DUF: Hierarchical Deep Generative Models for Design Under Free-Form Geometric Uncertainty

Abstract Deep generative models have demonstrated effectiveness in learning compact and expressive design representations that significantly improve geometric design optimization. However, these models do not consider the uncertainty introduced by manufacturing or fabrication. The past work that quantifies such uncertainty often makes simplifying assumptions on geometric variations, while the “real-world,” “free-form” uncertainty and its impact on design performance are difficult to quantify due to the high dimensionality. To address this issue, we propose a generative adversarial network-based design under uncertainty framework (GAN-DUF), which contains a deep generative model that simultaneously learns a compact representation of nominal (ideal) designs and the conditional distribution of fabricated designs given any nominal design. This opens up new possibilities of (1) building a universal uncertainty quantification model compatible with both shape and topological designs, (2) modeling free-form geometric uncertainties without the need to make any assumptions on the distribution of geometric variability, and (3) allowing fast prediction of uncertainties for new nominal designs. We can combine the proposed deep generative model with robust design optimization or reliability-based design optimization for design under uncertainty. We demonstrated the framework on two real-world engineering design examples and showed its capability of finding the solution that possesses better performance after fabrication.

Variational autoencoder-based topological optimization of an anechoic coating: An efficient- and neural network-based design

An anechoic coating is an artificial heterogeneous composite material composed of periodic cells with cavities. Using the local resonance of cavities to reduce their sound absorption frequencies and widen their frequency bands has been a research hotspot in recent years. One of the main challenges involves optimizing the cavity structure of an anechoic coating to obtain low-frequency, broadband, and strong sound absorption properties. In this paper, a variational autoencoder (VAE) model-based topology optimization method was investigated. First, the finite-element method (FEM) was used to calculate the sound absorption coefficient, and a dataset was constructed with samples whose average sound absorption coefficients ranged from 200 to 6000 Hz and were greater than 0.75. Then, the VAE model was trained to learn the key features of an anechoic coating. Finally, the data were reconstructed with Gaussian distributions. The decoder network of the trained VAE model was used to design a new anechoic coating. It took only approximately 3 s to generate 100 new topologies, and the average absorption coefficients were all greater than 0.754. This efficient neural network-based method can be further generalized to optimize the designs of various mechanical structural materials with specific functions.

Acoustic topology optimization using moving morphable components in neural network-based design

Acoustic topology optimization using moving morphable components in neural network-based design

Network-Based Design of Near-Infrared Lamb-Dip Experiments and the Determination of Pure Rotational Energies of H218O at kHz Accuracy

Taking advantage of the extreme absolute accuracy, sensitivity, and resolution of noise-immune-cavity-enhanced optical-heterodyne-molecular spectroscopy (NICE-OHMS), a variant of frequency-comb-assisted Lamb-dip saturation-spectroscopy techniques, the rotational quantum-level structure of both nuclear-spin isomers of H218O is established with an average accuracy of 2.5 kHz. Altogether, 195 carefully selected rovibrational transitions are probed. The ultrahigh sensitivity of NICE-OHMS permits the observation of lines with room-temperature absorption intensities as low as 10−27 cm molecule−1, while the superb resolution enables the detection of a doublet with a separation of only 286(17) kHz. While the NICE-OHMS experiments are performed in the near-infrared window of 7000–7350 cm−1, the lines observed allow the determination of all the pure rotational energies of H218O corresponding to J values up to 8, where J is the total rotational quantum number. Both network and quantum theory have been employed to facilitate the measurement campaign and the full exploitation of the lines resolved. For example, to minimize the experimental effort, the transitions targeted for observation were selected via the spectroscopic-network-assisted precision spectroscopy (SNAPS) scheme built upon the extended Ritz principle, the theory of spectroscopic networks, and an underlying dataset of quantum chemical origin. To ensure the overall connection of the ultraprecise rovibrational lines for both nuclear-spin isomers of H218O, the NICE-OHMS transitions are augmented with six accurate microwave lines taken from the literature. To produce absolute ortho-H218O energies, the lowest ortho energy is determined to be 23.754 904 61(19) cm−1. A reference, benchmark-quality line list of 1546 transitions, deduced from the ultrahigh-accuracy energy values determined in this study, provides calibration standards for future high-resolution spectroscopic experiments between 0–1250 and 5900–8380 cm−1.

RobustH∞Deep Neural Network-Based Filter Design of Nonlinear Stochastic Signal Systems

Robust<i>H_∞</i>Deep Neural Network-Based Filter Design of Nonlinear Stochastic Signal Systems

Network-based reference model tracking for vehicle turning with event-triggered transmission and interval communication delays

This paper deals with the network-based modelling and reference model tracking control of four-wheel-independent-drive electric vehicles for turning purpose. First, a reference model is designed by adjusting its input to characterise the expected neutral turning of the vehicle. By considering event-triggered transmission and interval communication delays, the network-based tracking control system attaining vehicle turning is modelled as an interval input delay system, where the input delay is bounded by the bounds of communication delays and the sampling period. Second, a new delay-dependent bounded real lemma with less conservatism is derived by proposing a matrix-based binary quadratic convex method, constructing a discontinuous augmented Lyapunov–Krasovskii functional tailored based on second-order Bessel–Legendre inequality, and exploiting the relation among the upper bounds of input delay and interval communication delays, and the sampling period. Third, the network-based tracking controller design result is further established by linear matrix inequalities. Finally, an example is provided to confirm the superiority of the analysis results and the effectiveness of control design.

LCL-T resonant network-based modular multi-channel constant-current LED driver analysis and design

Multiple output LED drivers are necessary to achieve better performance and higher reliability in street lighting, tunnel lighting and LCD background lighting applications. Based on LCL-T constant-current characteristics, a modular multiple output LED driver is proposed in this paper. The LCL-T rectifiers are connected to the same voltage bus and work like current sources. However, since there are much higher voltage harmonics in the AC square bus voltage, it is very important to quantitatively analyze how these harmonics influence the precision of the output currents. In addition, all of the switches can achieve Zero Voltage Switching (ZVS) by proper design of the LCL-T network. Finally, a 200-W prototype with five channels is established with an efficiency of 92.25%.

Neural network-based design and evaluation of performance metrics using adaptive line enhancer with adaptive algorithms for auscultation analysis

Auscultation is an important key for the physical (respiratory and circulatory) examination and is helpful in diagnosing various disorders. Auscultation is performed for the purposes of examining the circulatory and respiratory sounds and gastrointestinal system (bowel sounds). Besides inconsistencies in the propagation of the normal sounds, there are also several types of specific irregularities that can be heard in respiratory sounds: commonly known abnormal sounds in lung sound (wheezes, crackles, stridor, squawks, rhonchi and crackles) and heart sounds (heart murmurs). However, detection of abnormal sounds during auscultation needs extensive training and experience. Real-time separation of these heart sound signals from the lung sound signals is of great research interest and difficult to achieve. In this work, the authors proposed a novel adaptive line enhancer using nonlinear ANN design used for auscultation analysis. Our proposed designs are trained using different networks training algorithms which resulted in better mean square error reduction within compact time.

OBWD: an ontology and Bayesian network-based workflow design platform

OBWD: an ontology and Bayesian network-based workflow design platform