检索结果-内蒙古大学图书馆

Adaptive Simulation-Based Training of Artificial-Intelligence Decision Makers Using Bayesian optimization

JOURNAL OF AEROSPACE INFORMATION SYSTEMS 2018年第2期15卷 38-56页

作者： Israelsen, Brett Ahmed, Nisar Center, Kenneth Green, Roderick Bennett, Winston, Jr. Univ Colorado Comp Sci Boulder CO 80309 USA Univ Colorado Aerosp Engn Boulder CO 80309 USA Orbit Log Inc Greenbelt MD 20770 USA US Air Force Res Lab Warfighter Readiness Res Div Wright Patterson AFB OH 45433 USA

This work studies how an artifical-intelligence-controlled dogfighting agent with tunable decision-making parameters can learn to optimize performance against an intelligent adversary, as measured by a stochastic objective function evaluated on simulated combat engagements. Gaussian process Bayesian optimization techniques are developed to automatically learn global Gaussian process surrogate models, which provide statistical performance predictions in both explored and unexplored areas of the parameter space. This allows a learning engine to sample full-combat simulations at parameter values that are most likely to optimize performance and provide highly informative data points for improving future predictions. However, standard Gaussian process Bayesian optimization methods do not provide a reliable surrogate model for the highly volatile objective functions found in aerial combat and thus do not reliably identify global maxima. These issues are addressed by novel repeat sampling and hybrid repeat/multipoint sampling techniques. Simulation studies show that hybrid repeat/multipoint sampling improves the accuracy of Gaussian process surrogate models, allowing artificial-intelligence decision makers to more accurately predict performance and efficiently tune parameters.

关键词： Artificial Intelligence Inverse Reinforcement Learning Surrogate Model Probability Distribution Pilot optimization algorithm Black Box Numerical optimization Human Performance Modeling Human In The Loop

来源：评论

学校读者我要写书评

暂无评论

An optimal positioning algorithm for cumulative drag reduction in heterogeneous platoons 5

An optimal positioning algorithm for cumulative drag reducti...

引用

5th Indian Control Conference (ICC)

作者： Sahil, Kommalapati Mahindra Ecole Cent Hyderabad Telangana India

ISBN: (纸本)9781538662465

Positioning is a cardinal problem unique to heterogeneous platoons. The relative position of a vehicle in the platoon plays an important role in optimizing the net fuel consumption. The current work addresses this problem where a new algorithm is explored from an aerodynamically advantageous point of view to minimize the total drag experienced. It is further verified through numerical simulations on three different platoons.

关键词： Heterogeneous Platoons fuel reduction Air-foil Drag reduction optimization algorithm intelligent transport

来源：评论

学校读者我要写书评

暂无评论

Investigation of a Local Correlation-based Transition Model in a Newton-Krylov algorithm

Investigation of a Local Correlation-based Transition Model ...

引用

AIAA Science and Technology (SciTech) Forum and Exposition

作者： Piotrowski, Michael G. H. Zingg, David W. Univ Toronto Inst Aerosp Studies Toronto ON M3H 5T6 Canada NASA Ames Res Ctr Moffett Field CA 94035 USA

ISBN: (数字)9781624105784

ISBN: (纸本)9781624105784

A fully-coupled, implicit Newton-Krylov algorithm with laminar-turbulent boundary-layer transition prediction capabilities for three-dimensional external aerodynamic flows has been developed. The gamma-(R) over tildee(theta t) Local Correlation-Based Transition Model with helicity-based crossflow correlations is modified and fully coupled to the Spalart-Allmaras turbulence model in a parallel Newton-Krylov-Schur flow solver. Deep convergence is obtained using an efficient pseudo-transient continuation time marching strategy, consisting of an approximate-Newton phase followed by an inexact-Newton phase, with source-term time stepping and exponential penalty functions used to address the strong sources and non-smooth functions introduced by the transition model. A modification to the Flength function is introduced to improve the numerical convergence behaviour of the model. The current implementation is validated using two- and three-dimensional transition test cases, with the results obtained using the modified transition model comparing well with experiment. The elimination of discontinuities in the transition model makes it a suitable candidate for gradient-based shapea* optimization algorithms.

关键词： Solution algorithm Natural Laminar Flow Spalart Allmaras Turbulence Model Laminar Turbulent Transition optimization algorithm Angle of Attack Reynolds Averaged Navier Stokes Aerodynamic Shape optimization Swept Wing Free Stream Turbulence

来源：评论

学校读者我要写书评

暂无评论

Research on the optimization mechanism of loading path in hydroforming process

引用

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY 2018年第9-12期94卷 4125-4137页

作者： Feng, Ying-ying Luo, Zong-an Su, Hai-long Wu, Qing-lin Northeastern Univ State Key Lab Rolling & Automat Shenyang 110819 Liaoning Peoples R China

In this paper, the optimization mechanism of loading path in hydroforming process is researched. Firstly, the geometric model of an X-shaped tube is established by using 3D drawing software UG;the DYNAFORM software is used to simulate the forming performance of the X tube under different loading paths. The backward displacement is taken as the main factor of the loading path, and the loading path is presented in the form of three factor graphs;it can show the relationship of the main factors of the loading path visually, accurately, and precisely, which are axial feed, internal pressure, and back displacement. Secondly, the orthogonal test method is used to select the optimal loading path, and the back propagation (BP) neural network based on genetic algorithm is used to optimize the loading path of X tubes. Through synthetic consideration of the interrelation of the minimum wall thickness, the maximum wall thickness, the height of branch, and the contact area between branch tube and back punch, the average performance index function is established in the BP neural network control algorithm to optimize the learning efficiency and shorten the calculation time. Finally, verified by experiment, the optimization method of loading path for X tube hydroforming process could control the precision error of results between the simulation and the experiment within 5%, and has high accuracy and good feasibility.

关键词： Tube hydroforming Loading path optimization algorithm Average performance index function

来源：评论

学校读者我要写书评

暂无评论

High-to-Low Initial Sample Ratio of Hierarchical Kriging for Film Hole Array optimization

引用

JOURNAL OF PROPULSION AND POWER 2018年第1期34卷 108-115页

作者： Kim, Yoonki Lee, Sanga Yee, Kwanjung Rhee, Dong-Ho Seoul Natl Univ Dept Mech & Aerosp Engn Seoul 151742 South Korea Seoul Natl Univ Dept Mech & Aerosp Engn Inst Adv Aerosp Technol Seoul 151742 South Korea Korea Aerosp Res Inst Aeronaut Res & Dev Head Off Daejeon 305333 South Korea

This study aims at proving the existence of the optimal number of high initial sample for time-efficient optimization and demonstrating that the reliable optimization results can be guaranteed when the hierarchical kriging model is used as a surrogate model. The hierarchical kriging model is one sort of variable-fidelity modeling, which mimics the general tendency of a model based on the low-fidelity information and calibrates the accuracy of the model with the high-fidelity information. There, however, have been still several ambiguities to be clarified for practical utilization of hierarchical kriging model. For clarifying these, the film-cooling hole arrangement is selected as an optimization target. The optimization is progressed with the expected-improvement-based efficient globala* optimization algorithm coupled with the hierarchical kriging model. The fidelity in this paper is classified by mesh density: 2 million and 8 million grid system. According to this overall optimization process, case studies are conducted with respect to the number of high initial samples. As a result, the reliable optimization results are obtained regardless of different numbers of high initial samples, and approximately 40% time reduction can be obtained.

关键词： Kriging Model Richardson Extrapolation optimization algorithm Turbine Inlet Temperature Universal Gas Constant Reynolds Averaged Navier Stokes Mass Flow Rate CFD Adiabatic Conditions Rotor Aerodynamics

来源：评论

学校读者我要写书评

暂无评论

Example of a Pitfall in Aerodynamic Shape optimization

引用

AIAA JOURNAL 2018年第4期56卷 1532-1540页

作者： Destarac, Daniel Carrier, Gerald Anderson, George R. Nadarajah, Siva Poole, Daniel J. Vassberg, John C. Zingg, David W. ONERA French Aerosp Lab F-92190 Meudon France Sci & Technol Corp Moffett Field CA 94035 USA McGill Univ Montreal PQ H3A 2S6 Canada Univ Bristol Bristol BS8 1TR Avon England Boeing Co Long Beach CA 90808 USA Univ Toronto 100 Coll St Toronto ON M5S 1A1 Canada

The Applied Aerodynamics Technical Committee of AIAA launched an optimization discussion group, the Aerodynamic Design optimization Discussion Group, in 2013. One of the four benchmark test cases is based on the NACA 0012 airfoil and solutions of the Euler equations with prescribed objective function and geometric constaints. Volunteer participants were invited by ONERA-The French Aerospace Lab to submit their optimized shapes for uniform pressure drag assessment through grid convergence. The assessment required three phases. Initial investigations with a simple protocol proved insufficient and raised some questions. This led to further investigations, which showed that, for most optimized airfoils, there exists a lower drag branch and a higher drag branch in the drag (Mach number) diagram. Hysteresis was observed when performing a downward Mach number sweep and an upward Mach number sweep. For several airfoils, the design Mach number falls within the overlap range of the two branches, which indicates that two solutions may exist for the same airfoil. Two kinds of supersonic flow structure can be distinguished. One kind produces a lower drag branch and the other a higher drag branch. The jump from one branch to the other involves nonsymmetrical converged solutions. These further investigations required a final assessment, where a protocol was instituted to ensure that the lower drag branch was captured by taking advantage of the hysteresis. However, nonuniqueness may confound ana* optimization algorithm because the same set of design variables can give two different objective function values.

关键词： Aerodynamic Shape optimization NACA 0012 Supersonic Flow Hysteresis Euler Equations optimization algorithm Supersonic Wave Computational Fluid Dynamics Pressure Coefficient Finite Volume Formulation

来源：评论

学校读者我要写书评

暂无评论

Model Predictive Control Architectures for Maneuver Load Alleviation in Very Flexible Aircraft

Model Predictive Control Architectures for Maneuver Load All...

引用

AIAA Science and Technology (SciTech) Forum and Exposition

作者： Pereira, Mateus de F., V Kolmanovsky, Ilya Cesnik, Carlos E. S. Vetrano, Fabio Univ Michigan Dept Aerosp Engn Ann Arbor MI 48109 USA Airbus Operat SAS Loads & Aeroelast Flight Phys Toulouse France

ISBN: (数字)9781624105784

ISBN: (纸本)9781624105784

Two model predictive control (MPC) architectures to maneuver load alleviation in flexible and very flexible aircraft are considered. In the first, a tracking MPC formulation is used that directly manipulates the control inputs. In the second, MPC manipulates reference commands to the inner loop linear quadratic regulator (LQR) controllers. These MPC designs exploit full-order and reduced-order linearized models for prediction, and quadratic programming for computing the solutions. The performance of these control strategies is evaluated and compared through numerical simulations on nonlinear models of flexible and very flexible aircraft. Issues associated with the accuracy of the prediction model used in MPC design are discussed, particularly in view of the challenge of dealing with nonlinear very flexible aircraft dynamics.

关键词： Aeroplane Linear Quadratic Regulator optimization algorithm Aircraft Dynamics Numerical Simulation Computing High Altitude Long Endurance Euler Bernoulli Beams Wing Root Flight Control Surfaces

来源：评论

学校读者我要写书评

暂无评论

A human-computer cooperation improved ant colony optimization for ship pipe route design

引用

OCEAN ENGINEERING 2018年 150卷 12-20页

作者： Wang, Yun-long Yu, Yan-yun Li, Kai Zhao, Xue-guo Guan, Guan Dalian Univ Technol Sch Naval Architecture Dalian 116024 Peoples R China

This paper presents a human-computer cooperation improved ant colony optimization (HCCIACO) algorithm for ship pipe route design (SPRD). SPRD is a conbinatorial optimization problem with various performance constraints, its hard to fmd an effective solution only by computer. Based on the human-computer cooperation theory, the HCCIACO algorithm takes full advantage of designers' expertise and experience as well as computers' calculation ability. It conbines the artificial sulotion and algorithm solution in the genetic sense of the improved ant colony optimization (IACO) algorithm so that the optimization approach for SPRD in three-dimensional space can be obtained. The improved ant colony optimization simplifies the problem by reducing the complexity in calculation and engineering to some extent. Meanwhile, it guides the algorithm to search effectively for the stable solution which satisfies the engineering requirements. in this paper, the structure and updating method of artificial solution as well as the combination mode of artificial solution and algorithm solution have been researched. Compare with the conventional mathod, HCCIACO algorithm not only improves the convergence speed, but also improves the quality of the solution. Finally, the simulation results demonstrate the feasibility and efficiency of the proposed algorithm.

关键词： Ship pipe route design optimization algorithm Human-computer cooperation Artificial solution algorithm solution Ant colony algorithm

来源：评论

学校读者我要写书评

暂无评论

Aerodynamic Design of a Morphing Wing Sailplane

Aerodynamic Design of a Morphing Wing Sailplane

引用

AIAA Aviation Forum and Exposition

作者： Achleitner, Johannes Rohde-Brandenburger, Kai von Bieberstein, Paul Rogalla Sturm, Fabian Hornung, Mirko Tech Univ Munich Inst Aircraft Design Munich Germany DLR Inst Aerodynam & Flow Technol Braunschweig Germany

ISBN: (数字)9781624105890

ISBN: (纸本)9781624105890

The aerodynamic design methodology of a wing with a morphing forward section for an 18 meter class sailplane is shown in this paper. The airfoils have been designed by numerical optimization and reworked using inverse design methods. For the wing planform design a multiple vortex lifting line method has been coupled with ana* optimization algorithm. Overall aircraft performance calculations indicate the potential of the morphing forward wing section concept compared to conventional aircraft.

关键词： Lifting Line Method Transport Aircraft Lift to Drag Ratio Numerical optimization Wing Planforms optimization algorithm Aircraft Performance Vortices Aerodynamic Performance XFoil

来源：评论

学校读者我要写书评

暂无评论

Damage Precursor Identification in Composite Laminates using Data Driven Approach

Damage Precursor Identification in Composite Laminates using...

引用

AIAA Science and Technology (SciTech) Forum and Exposition

作者： Elenchezhian, Muthu Ram Prabhu Vadlamudi, Vamsee Raihan, Rassel Md Reifsnider, Kenneth L. Univ Texas Arlington Arlington TX 76019 USA Univ Texas Arlington Res Inst Ft Worth TX 76118 USA

ISBN: (数字)9781624105784

ISBN: (纸本)9781624105784

Composite materials are rapidly being used in commercial aviation and other day to day applications. The individual damage modes in composites are very well understood but it is the interaction of these local damage modes that leads to global failure. In the current research we intend to identify the damage precursors and the initiation of failure events in off axis unidirectional composite laminates loaded in quasi static uniaxial tension by measuring the dielectric response of the material by an in-situ technique using Broadband Dielectric Spectroscopy (BbDS). Using the variation of permittivity with strain, we are able to classify the stages of damage and predict the current material state. These data were then used to develop artificial intelligence models to identify the material state change and further use this data to predict the damage precursor stage and initiation of failure events. Different artificial intelligence models such as multi-layer perceptron, random forest regression and recurrent neural networks developed are discussed.

关键词： Fiber Reinforced Composites Convolutional Neural Network Dielectric Properties Artificial Intelligence Spectroscopy Commercial Aviation Stress Strain Curve Vanishing Gradient Problem optimization algorithm Electric Potential

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：