介绍了旋转体时域有限差分法(BOR-FDTD),导出了电磁场迭代计算公式。给出了自适应协方差矩阵进化策略(CMA-ES)的基本原理和步骤。提出了一种基于BOR-FDTD和CMA-ES的波纹喇叭优化设计技术,并将该项技术用于平方公里阵(Square Kilometre A...介绍了旋转体时域有限差分法(BOR-FDTD),导出了电磁场迭代计算公式。给出了自适应协方差矩阵进化策略(CMA-ES)的基本原理和步骤。提出了一种基于BOR-FDTD和CMA-ES的波纹喇叭优化设计技术,并将该项技术用于平方公里阵(Square Kilometre Array,SKA)望远镜天线Band 4(2.8~5.18 GHz)波纹喇叭馈源的优化设计。计算结果表明,该馈源在工作频带内反射损耗基本在-20 d B以下,天线口径效率均优于86.5%,且口径效率随频率的变化较小。展开更多
针对平方公里阵(Square Kilometre Array,SKA)天线对高灵敏度的需求,利用基于旋转体时域有限差分法(BOR-FDTD)和自适应协方差矩阵进化策略(CMA-ES)的波纹喇叭优化设计技术,提出了以灵敏度为目标的大张角波纹喇叭优化设计方法。分别以天...针对平方公里阵(Square Kilometre Array,SKA)天线对高灵敏度的需求,利用基于旋转体时域有限差分法(BOR-FDTD)和自适应协方差矩阵进化策略(CMA-ES)的波纹喇叭优化设计技术,提出了以灵敏度为目标的大张角波纹喇叭优化设计方法。分别以天线口径效率和灵敏度为优化目标对工作于4.6~8.51 GHz的大张角波纹喇叭进行优化设计。计算结果表明,以灵敏度为优化目标所设计的波纹喇叭综合性能更优,其交叉极化和反射损耗均优于-20 d B,用于SKA天线的口径效率在85.1%以上,灵敏度优于7.68 m^2/K。展开更多
研究模型选择对支持向量机(SVM)的泛化性能有着重要影响。针对传统梯度算法对初始值敏感及网格搜索法计算复杂的缺点,为了提高全面优化能力和分类精度,提出了一种基于协方差矩阵自适应进化策略(CMA-ES)的支持向量机(SVM)模型优化算法,...研究模型选择对支持向量机(SVM)的泛化性能有着重要影响。针对传统梯度算法对初始值敏感及网格搜索法计算复杂的缺点,为了提高全面优化能力和分类精度,提出了一种基于协方差矩阵自适应进化策略(CMA-ES)的支持向量机(SVM)模型优化算法,通过对SVM泛化性能界(Bounds on Generalization Performance)的优化求解,实现了基于CMA-ES算法的SVM模型选择。在标准数据集上的实验结果表明:相比遗传算法和梯度算法,上述方法能够在较小计算代价下得到更优的超参数,提高支持向量机的预测精度稳定性,尤其适合大样本数据条件下的模型选择。展开更多
Purpose – In the process of robot shell design, it is necessary to match the shape of the input 3D originalcharacter mesh model and robot endoskeleton, in order to make the input model fit for robot and avoidcollisio...Purpose – In the process of robot shell design, it is necessary to match the shape of the input 3D originalcharacter mesh model and robot endoskeleton, in order to make the input model fit for robot and avoidcollision. So, the purpose of this paper is to find an object of reference, which can be used for the process ofshape matching.Design/methodology/approach – In this work, the authors propose an interior bounded box (IBB)approach that derives from oriented bounding box (OBB). This kind of box is inside the closed mesh model.At the same time, it has maximum volume which is aligned with the object axis but is enclosed by all the meshvertices. Based on the IBB of input mesh model and the OBB of robot endoskeleton, the authors can completethe process of shape matching. In this paper, the authors use an evolutionary algorithm, covariance matrixadaptation evolution strategy (CMA-ES), to approximate the IBB based on skeleton and symmetry of inputcharacter mesh model.Findings – Based on the evolutionary algorithm CMA-ES, the optimal position and scale informationof IBB can be found. The authors can obtain satisfactory IBB result after this optimization process.The output IBB has maximum volume and is enveloped by the input character mesh model as well.Originality/value – To the best knowledge of the authors, the IBB is first proposed and used in the field ofrobot shell design. Taking advantage of the IBB, people can quickly obtain a shell model that fit for robot.At the same time, it can avoid collision between shell model and the robot endoskeleton.展开更多
为了解决目前水质预测中未考虑局部无知性这一问题,提出一种基于幂集置信规则库(Belief rule base with power set,PBRB)的水质预测模型。该模型能够有效融合专家知识与定量数据,并能在描述多种不确定性的同时,将传统的辨识框架扩展到幂...为了解决目前水质预测中未考虑局部无知性这一问题,提出一种基于幂集置信规则库(Belief rule base with power set,PBRB)的水质预测模型。该模型能够有效融合专家知识与定量数据,并能在描述多种不确定性的同时,将传统的辨识框架扩展到幂集,使其能够很好地表达无知性从而提高水质预测精度。此外,利用协方差矩阵自适应进化策略(Covariance matrix adaptive evolution strategy,CMA-ES)算法对PBRB模型进行优化。仿真结果表明:PBRB模型能准确预测一段时间内水质变化趋势,预测精度高于其他传统方法。展开更多
准确预测处理器性能对计算机硬件设计与改进有着重要意义。然而,处理器预测系统存在两个核心问题:预测过程中处理器内部构造复杂和不确定性以及预测结果的不可解释性。置信规则库作为一种基于IF-THEN规则的建模方法,具有一定的可解释性...准确预测处理器性能对计算机硬件设计与改进有着重要意义。然而,处理器预测系统存在两个核心问题:预测过程中处理器内部构造复杂和不确定性以及预测结果的不可解释性。置信规则库作为一种基于IF-THEN规则的建模方法,具有一定的可解释性并且可以处理复杂系统评估与预测中的不确定信息。但BRB的规则爆炸问题限制了专家知识的使用。因此,本文提出了一种基于近似置信规则库(ABRB)的处理器性能预测模型。该模型通过构建单属性BRB模型来解决规则爆炸问题,并通过基于投影协方差矩阵自适应进化策略(P-CMA-ES)算法对专家知识给定的初始参数进行优化。最后以UCI中处理器数据集为例,验证了所提方法的有效性。Accurate prediction of processor performance is important for computer hardware design and improvement. However, there are two core problems in processor prediction systems: the complexity and uncertainty of processor internals during the prediction process and the non-interpretability of the prediction results. Belief rule base (BRB), as a modelling method based on IF-THEN rules, has some interpretability and can handle uncertain information in the evaluation and prediction of complex systems. However, the rule explosion problem of BRB limits the use of expert knowledge. Therefore, this paper proposes a processor performance prediction model based on approximate belief rule base. The model solves the rule explosion problem by constructing a single-attribute BRB model and optimizes the initial parameters given by the expert knowledge by the Projection Covariance Matrix Adaptive Evolutionary Strategy (P-CMA-ES) based algorithm. Finally, the effectiveness of the proposed method is validated using the UCI mid-processor dataset as an example.展开更多
文摘介绍了旋转体时域有限差分法(BOR-FDTD),导出了电磁场迭代计算公式。给出了自适应协方差矩阵进化策略(CMA-ES)的基本原理和步骤。提出了一种基于BOR-FDTD和CMA-ES的波纹喇叭优化设计技术,并将该项技术用于平方公里阵(Square Kilometre Array,SKA)望远镜天线Band 4(2.8~5.18 GHz)波纹喇叭馈源的优化设计。计算结果表明,该馈源在工作频带内反射损耗基本在-20 d B以下,天线口径效率均优于86.5%,且口径效率随频率的变化较小。
文摘针对平方公里阵(Square Kilometre Array,SKA)天线对高灵敏度的需求,利用基于旋转体时域有限差分法(BOR-FDTD)和自适应协方差矩阵进化策略(CMA-ES)的波纹喇叭优化设计技术,提出了以灵敏度为目标的大张角波纹喇叭优化设计方法。分别以天线口径效率和灵敏度为优化目标对工作于4.6~8.51 GHz的大张角波纹喇叭进行优化设计。计算结果表明,以灵敏度为优化目标所设计的波纹喇叭综合性能更优,其交叉极化和反射损耗均优于-20 d B,用于SKA天线的口径效率在85.1%以上,灵敏度优于7.68 m^2/K。
文摘研究模型选择对支持向量机(SVM)的泛化性能有着重要影响。针对传统梯度算法对初始值敏感及网格搜索法计算复杂的缺点,为了提高全面优化能力和分类精度,提出了一种基于协方差矩阵自适应进化策略(CMA-ES)的支持向量机(SVM)模型优化算法,通过对SVM泛化性能界(Bounds on Generalization Performance)的优化求解,实现了基于CMA-ES算法的SVM模型选择。在标准数据集上的实验结果表明:相比遗传算法和梯度算法,上述方法能够在较小计算代价下得到更优的超参数,提高支持向量机的预测精度稳定性,尤其适合大样本数据条件下的模型选择。
基金This research,which is carried out at BeingThere Centre,collaboration among IMI of Nanyang Technological University(NTU)Singapore,ETH Zurich and UNC Chapel Hill,is supported by the Singapore National Research Foundation(NRF)under its International Research Centre@Singapore Funding Initiative and administered by the Interactive Digital Media Programme Office(IDMPO).The author Shihui Guo is supported by Chinese Post-doctoral Science Foundation 2016M600506.
文摘Purpose – In the process of robot shell design, it is necessary to match the shape of the input 3D originalcharacter mesh model and robot endoskeleton, in order to make the input model fit for robot and avoidcollision. So, the purpose of this paper is to find an object of reference, which can be used for the process ofshape matching.Design/methodology/approach – In this work, the authors propose an interior bounded box (IBB)approach that derives from oriented bounding box (OBB). This kind of box is inside the closed mesh model.At the same time, it has maximum volume which is aligned with the object axis but is enclosed by all the meshvertices. Based on the IBB of input mesh model and the OBB of robot endoskeleton, the authors can completethe process of shape matching. In this paper, the authors use an evolutionary algorithm, covariance matrixadaptation evolution strategy (CMA-ES), to approximate the IBB based on skeleton and symmetry of inputcharacter mesh model.Findings – Based on the evolutionary algorithm CMA-ES, the optimal position and scale informationof IBB can be found. The authors can obtain satisfactory IBB result after this optimization process.The output IBB has maximum volume and is enveloped by the input character mesh model as well.Originality/value – To the best knowledge of the authors, the IBB is first proposed and used in the field ofrobot shell design. Taking advantage of the IBB, people can quickly obtain a shell model that fit for robot.At the same time, it can avoid collision between shell model and the robot endoskeleton.
文摘为了解决目前水质预测中未考虑局部无知性这一问题,提出一种基于幂集置信规则库(Belief rule base with power set,PBRB)的水质预测模型。该模型能够有效融合专家知识与定量数据,并能在描述多种不确定性的同时,将传统的辨识框架扩展到幂集,使其能够很好地表达无知性从而提高水质预测精度。此外,利用协方差矩阵自适应进化策略(Covariance matrix adaptive evolution strategy,CMA-ES)算法对PBRB模型进行优化。仿真结果表明:PBRB模型能准确预测一段时间内水质变化趋势,预测精度高于其他传统方法。
文摘准确预测处理器性能对计算机硬件设计与改进有着重要意义。然而,处理器预测系统存在两个核心问题:预测过程中处理器内部构造复杂和不确定性以及预测结果的不可解释性。置信规则库作为一种基于IF-THEN规则的建模方法,具有一定的可解释性并且可以处理复杂系统评估与预测中的不确定信息。但BRB的规则爆炸问题限制了专家知识的使用。因此,本文提出了一种基于近似置信规则库(ABRB)的处理器性能预测模型。该模型通过构建单属性BRB模型来解决规则爆炸问题,并通过基于投影协方差矩阵自适应进化策略(P-CMA-ES)算法对专家知识给定的初始参数进行优化。最后以UCI中处理器数据集为例,验证了所提方法的有效性。Accurate prediction of processor performance is important for computer hardware design and improvement. However, there are two core problems in processor prediction systems: the complexity and uncertainty of processor internals during the prediction process and the non-interpretability of the prediction results. Belief rule base (BRB), as a modelling method based on IF-THEN rules, has some interpretability and can handle uncertain information in the evaluation and prediction of complex systems. However, the rule explosion problem of BRB limits the use of expert knowledge. Therefore, this paper proposes a processor performance prediction model based on approximate belief rule base. The model solves the rule explosion problem by constructing a single-attribute BRB model and optimizes the initial parameters given by the expert knowledge by the Projection Covariance Matrix Adaptive Evolutionary Strategy (P-CMA-ES) based algorithm. Finally, the effectiveness of the proposed method is validated using the UCI mid-processor dataset as an example.
