High-fidelity Evaluation Research Articles

Latent Crossover for Data-Driven Multifidelity Topology Design

Abstract Topology optimization is one of the most flexible structural optimization methodologies. However, in exchange for its high level of design freedom, typical topology optimization cannot avoid multimodality, where multiple local optima exist. This study focuses on developing a gradient-free topology optimization framework to avoid being trapped in undesirable local optima. Its core is a data-driven multifidelity topology design (MFTD) method, in which the design candidates generated by solving low-fidelity topology optimization problems are updated through a deep generative model and high-fidelity evaluation. As its key component, the deep generative model compresses the original data into a low-dimensional manifold, i.e., the latent space, and randomly arranges new design candidates over the space. Although the original framework is gradient free, its randomness may lead to convergence variability and premature convergence. Inspired by a popular crossover operation of evolutionary algorithms (EAs), this study merges the data-driven MFTD framework and proposes a new crossover operation called latent crossover. We apply the proposed method to a maximum stress minimization problem in 2D structural mechanics. The results demonstrate that the latent crossover improves convergence stability compared to the original data-driven MFTD method. Furthermore, the optimized designs exhibit performance comparable to or better than that in conventional gradient-based topology optimization using the P-norm measure.

High-Performance Uniform Damping Model for Response History Analysis in OpenSees

ABSTRACT A realistic and practical damping model is essential for structural dynamics. In this study, the second-order accurate time discretization of a high-performance uniform damping model for response history analysis is derived. Its stability and accuracy are proven, and its object-oriented implementation is introduced in OpenSees. The proposed time discretization accurately represents energy dissipation in individual components with minimal additional computational costs. It applies to various time integrators, element formulations, and structural systems, allowing for high-fidelity evaluation of structural performance and ensuring public safety. These advantages are illustrated through detailed case studies.

Data-driven multifidelity topology design using a deep generative model: Application to forced convection heat transfer problems

Topology optimization is a powerful methodology for generating novel designs with a high degree of design freedom. In exchange for this attractive feature, topology optimization cannot generally avoid multimodality, which often impedes finding a satisfactory solution when dealing with strongly nonlinear optimization problems. In this study, we focus on constructing a framework that aims to indirectly solve such complex topology optimization problems. The framework is based on multifidelity topology design (MFTD), the basic concept of which is to divide solving an original topology optimization problem into two kinds of procedures, i.e., low-fidelity optimization and high-fidelity evaluation. We build the framework as a data-driven approach, where the design candidates given by solving the low-fidelity optimization problem are iteratively updated based on the idea of the evolutionary algorithm (EA). As a key procedure to realize a crossover-like operation in the high-dimensional design space, a variational autoencoder—one of the representative deep generative models—is utilized for generating a new dataset composed of various material distributions. Besides, we propose a mutation-like operation for generating novel material distributions based on reference ones in the dataset. By integrating these operations with an elitism-based selection procedure, we propose data-driven MFTD that enables gradient-free optimization even if tackling a complex optimization problem with a high degree of design freedom. We apply the proposed framework to forced convection heat transfer problems, where the low-fidelity optimization problem is formulated using a Darcy flow model, whereas the high-fidelity evaluation is conducted using a Navier–Stokes model. We first demonstrate that the obtained results from the proposed framework can achieve almost identical performance compared with that of the directly solved results in a well-known 2D laminar heat transfer problem. We then show that the proposed framework can find solutions in minimax problems of topology optimization pertaining to 3D laminar and turbulent heat transfer.

Transfer stacking from low-to high-fidelity: A surrogate-assisted bi-fidelity evolutionary algorithm

Optimization of many real-world optimization problems relies on numerical simulations for function evaluations. In some cases, both high- and low-fidelity simulations are available, where the high fidelity evaluation is accurate but time-consuming, whereas the low-fidelity evaluation is less accurate but computationally cheap. To find an acceptable optimum within a limited budget, it is economical for evolutionary algorithms to use both high- and low-fidelity evaluations in a single optimization search. This paper proposes a novel surrogate-assisted evolutionary algorithm using the transfer stacking technique for bi-fidelity optimization. To this end, a radial basis function network is firstly built to approximate the high-fidelity fitness function as additional low-fidelity evaluation, then a surrogate model transferring the original and additional low-fidelity evaluations to the expensive high-fidelity evaluation is adapted to guide the search. The simulation results on a series of bi-fidelity optimization benchmark problems with resolution, stochastic, and instability errors and a beneficiation processes optimization problem show that the proposed algorithm is both effective and efficient for solving bi-fidelity optimization problems, when their low-fidelity evaluations have resolution and stochastic errors.

An Ontology-Based Predictive Maintenance Tool for Power Substation Faults in Distribution Grid

Recent advances in Power Grid (PG) technology pose an important problem of measuring the effectiveness of power grid configurations. Current assessment models are not adequate to mitigate the setup issues due to the absence of a high-fidelity evaluation framework that can consider diverse scenarios based on the market interest. Consequently, we develop a highly flexible Ontology-based Evaluation System that can accommodate and assess different scenarios. The use of ontology as middleware is the best approach to produce an efficient, semantically aware, and operationally accurate system environment for managing flexibility in evaluation. The evaluation is made by predicting the failure intensity and subsequently generate a maintenance report of a particular configuration. The selection of the best configuration is made by comparing the maintenance report of different configurations. The developed evaluation system consists of three main components which are Configuration Generator Tool (GCT), Failure Prediction Model (FDM), and Hybrid Simulation Platform (HSP). The GCT is a knowledge-based system that provides a powerful tool for engineers to generate alternative configurations. The GCT data were collected from literature, validated by experts, and modeled using Web Ontology Language (OWL). While the HSP was developed using several modelings and ontology-based tools such as blender 3D modeling, unity 3d, asp.net, my sql, and apache Jena fuseki. Finally, the FDM was developed based on the impact and relationship of odd events to power grid components and the impact of a failed component to other components, the prediction is modeled using two methods Poisson Model and Likelihood Estimation Method.

Performance bookkeeping for aircraft configurations with fuselage wake-filling propulsion integration

The aim of this research is the identification of a unified bookkeeping and evaluation scheme for the integrated performance analysis of a boundary layer ingesting (BLI) concept in the conceptual design phase. A thorough review and classification of existing performance bookkeeping schemes suits as a basis for the derivation of a bookkeeping scheme suitable for the initial sizing as well as detailed design analysis during the conceptual phase of a BLI concept. Figures of merit for the concept performance assessment are evaluated with regard to the requirements of aircraft multidisciplinary conceptual design. Based on the survey, the most practical integral momentum conservation approach is deduced and its application to integrated conceptual sizing and a subsequent design analysis is evaluated. The proposed scheme is universally applicable to coupled airframe–propulsion aircraft concepts, compatible with standard aircraft and propulsion system sizing tools and, under certain assumptions, deployable for low- and high-fidelity evaluation methods. Finally, several figures of merit are selected to cover a range of design aspects in the BLI evaluation.

Multifidelity design guided by topology optimization

In this study, we present a framework based on the concept of multifidelity design optimization with the purpose of indirectly solving complex—computationally heavy and/or unstable—topology optimization problems. Our primary idea is to divide an original topology optimization problem into two types of subproblems for low-fidelity optimization and high-fidelity evaluation. To realize this idea, artificial design parameters, which we refer to as seeding parameters, are incorporated into the low-fidelity optimization problem for generating various patterns of topology-optimized candidates. The low-fidelity optimization problem is deliberately formulated as an easily solvable one by decreasing the nonlinearity of the original physical phenomena. Notably, selecting valid seeding parameters in the low-fidelity optimization problem is essential for employing the proposed framework. The aim of high-fidelity evaluation is to obtain a satisfactory solution using a high-fidelity analysis model, which considers the nonlinearity of the original physical phenomena, from the data set of the topology-optimized candidates, via the design of experiments. We apply the proposed framework to two case studies of isothermal and thermal turbulent flow problems, and discuss its efficacy as an alternative strategy for solving complex topology optimization problems.

Efficient Design Optimization Assisted by Sequential Surrogate Models

The paper proposes a global optimization algorithm employing surrogate modeling and adaptive infill criteria. The surrogates are exploited to screen the design space and provide lower-fidelity predictions across it; on the other hand, specific criteria are designed to suggest new points for high-fidelity evaluation so as to enrich the optimizer database. Both Kriging and radial basis function network are used as surrogates with different training strategies. Sequential design is achieved by introducing several infill criteria according to the realization of the exploration-exploitation trade-off. Optimization results are provided both for scalable and analytical test functions and for a practical aerodynamic shape optimization problem.

A multi-objective optimization framework for robust axial compressor airfoil design

Airfoil design for stationary gas turbines is a challenging task involving both aerodynamic and structural aspects. The paper describes a multidisciplinary optimization process for axial compressor airfoils which is able to find optimal designs w.r.t. multiple objectives and constraints starting from a reference design and very few specifications of the new compressor. The process allows to simultaneously execute arbitrarily many instances of design evaluation processes independently from each other, which speeds it up, not just due to parallelization, but also because fast-running low-fidelity evaluation may take the design lead at an early design stage, whereas high-fidelity evaluation processes simultaneously contribute with more reliable results on the actual performance. For consistency of aerodynamic and structural analysis, an innovative method for direct loaded-to-unloaded design transformation is incorporated. Additionally, the process accounts for design robustness by utilizing production tolerances as an optimization objective. Therefore, a procedure is developed which allows to find the production tolerance which may be allowed without violating any constraints. An application example demonstrates that the proposed optimization process incorporating automatic detection of failure-critical eigenmode bands is able to shift them such that structurally reliable, robust, and simultaneously aerodynamically efficient designs are obtained.

Hybrid surrogate-model-based multi-fidelity efficient global optimization applied to helicopter blade design

ABSTRACTA multi-fidelity optimization technique by an efficient global optimization process using a hybrid surrogate model is investigated for solving real-world design problems. The model constructs the local deviation using the kriging method and the global model using a radial basis function. The expected improvement is computed to decide additional samples that can improve the model. The approach was first investigated by solving mathematical test problems. The results were compared with optimization results from an ordinary kriging method and a co-kriging method, and the proposed method produced the best solution. The proposed method was also applied to aerodynamic design optimization of helicopter blades to obtain the maximum blade efficiency. The optimal shape obtained by the proposed method achieved performance almost equivalent to that obtained using the high-fidelity, evaluation-based single-fidelity optimization. Comparing all three methods, the proposed method required the lowest total number of high-fidelity evaluation runs to obtain a converged solution.

Optimal Scheduling and Management on Correlating Reliability, Performance, and Energy Consumption for Multiagent Cloud Systems

In large-scale cloud computing environments, a multiagent cloud system (MACS) can be developed for effectively managing cloud infrastructure. In principle, a MACS has two essential components, a global agent (GA) and multiple local agents (LAs). The GA dispatches user requests to multiple LAs, while the LAs make optimal resource management strategies for serving user requests received. To develop optimal request scheduling strategies and resource management strategies for the MACS, multiple correlated metrics, particularly reliability, performance, and energy consumption need to be comprehensively considered. In this paper, a reliability-performance-energy correlation model is first proposed. The new model captures significant effects of random resource failures and recovery on MACS performance and energy consumption to ensure high fidelity and precise evaluation. A net profit optimization model depicting the tradeoff between performance and energy consumption is then developed based on the proposed theoretical model. Furthermore, an approach for optimization is proposed, where LAs can perform dynamic resource management for reaching optimum in the net profit metric. A genetic algorithm is also designed and implemented for searching the solution of global request scheduling for the GA. Numerical examples illustrate that the MACS can achieve efficient and comprehensive scheduling and management capability based on cooperation between the GA and LAs.

Redesign and Validation of Sisom, an Interactive Assessment and Communication Tool for Children With Cancer.

BackgroundChildren with cancer undergo intensive and long treatment periods that expose them and their families to a number of difficult physical, mental, and social challenges. Empowering children by actively involving them in their care can help them to cope with these challenges. It can, however, be difficult for children to be involved and talk about their illness experiences in a “traditional” conversation with health care professionals, especially for younger children. Sisom (Norwegian acronym “Si det som det er” or “Tell it how it is”) is an interactive computer-based assessment and communication tool to give children (aged 6-12 years) with cancer a “voice” in their care. Because of technological advances and widespread use of mobile devices Sisom had to be redesigned to better meet the needs of children of today.ObjectiveTo redesign Sisom for use on mobile devices and to validate and adapt it for use in a Swedish population of children with cancer.MethodsA user-experience design was used. Content adaptation included forward-backward translation by Swedish and Norwegian translators. Healthy children (n=5), children with experiences of cancer treatment (n=5) and their parents (n=5), and pediatric nurses (n=2) were then involved in culturally adapting Sisom to the Swedish context. The iterative low- and high-fidelity evaluation was supported by a think aloud method, semistructured interviews, and drawings to capture children’s views of Sisom. The redesign and evaluation continued until no further changes or improvements were identified by the participants or the researchers.ResultsChildren, parents, and pediatric nurses offered many suggestions for improvements to the original version in terms of content, aesthetics, and usability of Sisom. The most significant change that emerged through user input was a modification that entailed not using problem-focused statements in the assessment items. The parents and pediatric nurses considered the revised assessment items to be general and less diagnosis specific. The evaluation of aesthetics resulted in brighter colors and more positive and exciting details in the animations. The evaluation of usability included improvements of the verbal instructions on how to navigate in Sisom 2, and also that the answers to assessment items in Sisom 2 should be saved to provide the children with the option to pause and to continue answering the remaining assessment items at a later stage.ConclusionsOverall, this paper describes the process of using user-experience design with children in order to redesign and validate an interactive assessment and communication tool and how the outcomes of this process resulted in a new version, Sisom 2. All participants confirmed the usability and qualities of using the final version. Future research should be directed toward the implementation of Sisom 2 in clinical practice and to evaluate outcomes from individual and organizational levels.

Real-time computer-generated integral imaging and 3D image calibration for augmented reality surgical navigation

Autostereoscopic 3D image overlay for augmented reality (AR) based surgical navigation has been studied and reported many times. For the purpose of surgical overlay, the 3D image is expected to have the same geometric shape as the original organ, and can be transformed to a specified location for image overlay. However, how to generate a 3D image with high geometric fidelity and quantitative evaluation of 3D image's geometric accuracy have not been addressed. This paper proposes a graphics processing unit (GPU) based computer-generated integral imaging pipeline for real-time autostereoscopic 3D display, and an automatic closed-loop 3D image calibration paradigm for displaying undistorted 3D images. Based on the proposed methods, a novel AR device for 3D image surgical overlay is presented, which mainly consists of a 3D display, an AR window, a stereo camera for 3D measurement, and a workstation for information processing. The evaluation on the 3D image rendering performance with 2560×1600 elemental image resolution shows the rendering speeds of 50–60 frames per second (fps) for surface models, and 5–8 fps for large medical volumes. The evaluation of the undistorted 3D image after the calibration yields sub-millimeter geometric accuracy. A phantom experiment simulating oral and maxillofacial surgery was also performed to evaluate the proposed AR overlay device in terms of the image registration accuracy, 3D image overlay accuracy, and the visual effects of the overlay. The experimental results show satisfactory image registration and image overlay accuracy, and confirm the system usability.

Use of the Backseat Driving Technique in Evaluation of a Perceptually Optimized In-Car Navigation Display

Recently drivers have greatly benefited from new vehicular technologies such as in-car navigation systems, but at the same time they can be easily distracted from those technologies. Consequently, creating displays that balance the communication of information with the attentional demand imposed on the driver is of increasing importance. A set of research has been conducted to minimize the driver’s attentional cost toward the navigational displays while driving. However, accurate evaluation methods to assess the effects of these displays in realistic environments are not yet available. This article introduces the backseat driving technique for high-fidelity, safe, and inexpensive evaluation of interaction with in-car displays. This technique makes use of a real vehicle driving on a real road. As a result it allows for the exploration of some types of research questions using audio, visual, and kinesthetic stimulus at least equal in fidelity to very high-end driving simulators, without requiring such a specialized and expensive facility. Further, it allows for the employment of detailed and fine-grained measures of attentional demand, which cannot be safely used with subjects who are actually driving. Although the backseat driving technique can address only some types of questions and so is not a full replacement for high-realism driving simulators, it may offer a new approach, which augments laboratory, simulator, and real driving for many studies. As a part of the work presented here, the backseat driving technique is used to evaluate a previously developed in-car navigation display—the MOVE system. The technique allowed for new questions to be asked, which were not able to be considered in previous laboratory studies, and for the use of study measures that were only previously able to be used in the laboratory due to driving safety concerns. Specifically, the display was shown to work well when real-world stimulus are used to navigate along a real route, reducing the navigation error rate nearly threefold, and up to sixfold when compared to displays providing more or less contextual information. In addition the display was shown to cut total display fixation time (which is time spent looking away from the road) almost in half in both cases.

QCM-based measurement of chlorine-induced polymer degradation kinetics.

Highly structured network polymers are prepared via a molecular layer by layer technique (mLbL) and used as a model system to study aqueous degradation of polymer thin films in real time. Quantitative analysis of the degradation kinetics was enabled by the use of a quartz crystal microbalance (QCM). We conclude that the common metric of halogen, specifically chlorine, exposure (concentration × time) to be an ineffective normalization unit and showed a multistage adsorption process consistent with the established chemical mechanism. Additionally, degradation progression was tracked at multiple points of exposure to determine the effects of chlorination on the chemical and morphological state of the polymer structure with X-ray photoelectron spectroscopy (XPS) and atomic force microscopy (AFM), respectively. The formation of known halogenation products were corroborated with XPS through the high resolution spectra. Insight into the heterogeneous nature of the nanostructural degradation was derived from the AFM images. Periodic rinsing was found to release adsorbed chlorine but had negligible benefits on extending the exposure limits of the polyamide film. Fluorinated amine monomer (3,4-difluoroaniline) was incorporated into the surface of the polymer to determine the effect of limiting N-halogenation and the formation of the halogenated ring product. The modified surface layer reduced the rate and magnitude of chlorine adsorption relative to the neat polyamide surface. The QCM technique was shown to be an effective tool for rapid and high fidelity evaluation of molecular degradation and modification strategies to increase device lifetimes.

Updating and Extending the IRDF-2002 Dosimetry Library

Abstract The International Reactor Dosimetry File (IRDF)-2002 released in 2004 by the IAEA (see http://www-nds.iaea.org/irdf2002/) contains cross-section data and corresponding uncertainties for 66 dosimetry reactions. New cross-section evaluations have become available recently that re-define some of these dosimetry reactions including: (1) high-fidelity evaluation work undertaken by one of the authors (KIZ); (2) evaluations from the US ENDF/B-VII.0 and candidate evaluations from the US ENDF/B-VII.1 libraries that cover reactions within the International Evaluation of Neutron Cross-Section Standards; (3) European JEFF3.1 library; and (4) Japanese JENDL-4.0 library. Additional high-threshold reactions not included in IRDF-2002 (e.g.,59Co(n,3n) and 209Bi(n,3n)) have been also evaluated to characterize higher-energy neutron fields. Overall, 37 new evaluations of dosimetry reactions have been assessed and intercomparisons made with integral measurements in reference neutron fields to determine whether they should be adopted to update and improve IRDF-2002. Benchmark calculations performed for newly evaluated reactions using the ENDF/B-VII.0 235U thermal fission and 252Cf spontaneous fission neutron spectra show that calculated integral cross sections exhibit improved agreement with evaluated experimental data when compared with the equivalent data from the IRDF-2002 library. Data inconsistencies or deficiencies of new evaluations have been identified for 63Cu(n,2n), 60Ni(n,p)60m+g Co, 55Mn(n,γ), and 232Th(n,f) reactions. Compared with IRDF-2002, the upper neutron energy boundary was formally increased from the actual maximum energy of typically 20 MeV up to 60 MeV by using the TENDL-2010 cross sections and covariance matrices. This extension would allow the updated IRDF library to be also used in fusion dosimetry applications. Uncertainties in the cross sections for all new evaluations are given in the form of relative covariance matrices. Newly evaluated excitation functions should be considered as suitable candidates in the preparation of an improved version of the IRDF that was planned to be released for testing in December 2011.

Measurement of Temporal Awareness in Air Traffic Control

Temporal awareness, or level 3 situation awareness, is critical to successful control of air traffic, yet the construct remains ill-defined and difficult to measure. This research sought evidence for air traffic controllers' awareness of temporal characteristics of their tasks in data from a high-fidelity system evaluation simulation. Five teams of controllers worked on four scenarios with different traffic load. Several temporal parameters were defined for each task controllers performed during a simulation run and their actions on the tasks were timed relative to them. Controllers showed a strong tendency to prioritize tasks according to a first come, first served principle. This trend persisted as task load increased. Also evident was awareness of the urgency of tasks, as tasks with impending closing of a window of opportunity were performed before tasks that had longer time available before closing of the window.

Knowledge Discovery for Transonic Regional-Jet Wing through Multidisciplinary Design Exploration

Data mining is an important facet of solving multi-objective optimization problem. Because it is one of the effective manner to discover the design knowledge in the multi-objective optimization problem which obtains large data. In the present study, data mining has been performed for a large-scale and real-world multidisciplinary design optimization (MDO) to provide knowledge regarding the design space. The MDO among aerodynamics, structures, and aeroelasticity of the regional-jet wing was carried out using high-fidelity evaluation models on the adaptive range multi-objective genetic algorithm. As a result, nine non-dominated solutions were generated and used for tradeoff analysis among three objectives. All solutions evaluated during the evolution were analyzed for the tradeoffs and influence of design variables using a self-organizing map to extract key features of the design space. Although the MDO results showed the inverted gull-wings as non-dominated solutions, one of the key features found by data mining was the non-gull wing geometry. When this knowledge was applied to one optimum solution, the resulting design was found to have better performance compared with the original geometry designed in the conventional manner.

Multidisciplinary Design Optimization and Data Mining for Transonic Regional-Jet Wing

A large-scale, real-world application of evolutionary multi-objective optimization is reported. The multidisciplinary design optimization among aerodynamics, structures, and aeroelasticity of the wing of a transonic regional-jet aircraft was performed using high-fidelity evaluation models. Euler and Navier-Stokes solvers were employed for aerodynamic evaluation. The commercial software NASTRAN was coupled with a computational fluid dynamics solver for the structural and aeroelastic evaluations. An adaptive range multi-objective genetic algorithm was employed as an optimizer. The objective functions were minimizations of block fuel and maximum takeoff weight in addition to drag divergence between transonic and subsonic flight conditions. As a result, nine nondominated solutions were generated and used for tradeoff analysis among three objectives. Moreover, all solutions evaluated during the evolution were analyzed using a self-organizing map as a data mining technique to extract key features of the design space. One of the key features found by data mining was the nongull wing geometry, although the present multidisciplinary design optimization results showed the inverted gull wings as nondominated solutions. When this knowledge was applied to one optimum solution, the resulting design was found to have better performance and to achieve 3.6% improvement in the block fuel compared to the original geometry designed in the conventional manner.

Tumoral and non-tumoral trachea stenoses: evaluation with three-dimensional CT and virtual bronchoscopy

BackgroundWe evaluated the ability of 3D-CT and virtual bronchoscopy to estimate trachea stenosis in comparison to conventional axial CT and fiberoptic bronchoscopy, with a view to assist thoracic surgeons in depicting the anatomical characteristics of tracheal strictures.MethodsSpiral CT was performed in 16 patients with suspected tracheal stenoses and in 5 normal subjects. Tracheal stenoses due to an endoluminal neoplasm were detected in 13 patients, whilst post-intubation tracheal stricture was observed in the other 3 patients. Multiplanar reformatting (MPR), volume rendering techniques (VRT) and virtual endoscopy (VE) for trachea evaluation were applied and findings were compared to axial CT and fiberoptic bronchoscopy. The accuracy of the procedure in describing the localization and degree of stenosis was tested by two radiologists in a blinded controlled trial.ResultsThe imaging modalities tested showed the same stenoses as the ones detected by flexible bronchoscopy and achieved accurate and non-invasive morphological characterization of the strictures, as well as additional information about the extraluminal extent of the disease. No statistically significant difference was observed between the bronchoscopic findings and the results of axial CT estimations (P = 1.0). No statistically significant differences were observed between bronchoscopic findings and the MPR, VRT and VE image evaluations (P = 0.705, 0.414 and 0.414 respectively).ConclusionCT and computed generated images may provide a high fidelity, noninvasive and reproducible evaluation of the trachea compared to bronchoscopy. They may play a role in assessment of airway patency distal to high-grade stenoses, and represent a reliable alternative method for patients not amenable to conventional bronchoscopy.