With the rapid development of the cloud computing technology, it has matured enough for a lot of individuals and organizations to move their work into the cloud. Correspondingly, a variety of cloud services are emergi...With the rapid development of the cloud computing technology, it has matured enough for a lot of individuals and organizations to move their work into the cloud. Correspondingly, a variety of cloud services are emerging. It is a key issue to assess the cloud services in order to help the cloud users select the most suitable cloud service and the cloud providers offer this service with the highest quality. The criteria parameters defining the cloud services are complex which lead to cloud service deviation. In this paper, we propose an assessment method of parameters importance in cloud services using rough set theory. The method can effectively compute the importance of cloud services parameters and sort them. On the one hand, the calculation can be used as the credible reference when users choose their appropriate cloud services. On the other hand, it can help cloud service providers to meet user requirements and enhance the user experience. The simulation results show the effectiveness of the method and its relevance in the cloud context.展开更多
针对多目标识别过程中点云分类和分割精度不高的问题,提出了一种基于改进Transformer模型的点云分类与分割方法DRPT(Double randomness Point Transformer),该方法在Transformer模型卷积投影层创建新的点嵌入,利用局部邻域的动态处理在...针对多目标识别过程中点云分类和分割精度不高的问题,提出了一种基于改进Transformer模型的点云分类与分割方法DRPT(Double randomness Point Transformer),该方法在Transformer模型卷积投影层创建新的点嵌入,利用局部邻域的动态处理在数据特征向量中持续增加全局特征属性,从而提高多目标识别中点云分类和分割的精度。实验中采用了标准基准数据集(ModelNet40、ShapeNet部分分割和SemanticKITTI场景语义分割数据集)以验证模型的性能,实验结果表明:DRPT模型的pIoU值为85.9%,比其他模型平均高出3.5%,有效提高了多目标识别检测时点云分类与分割精度,是对智能网联技术发展的有效支撑。展开更多
参数化水平集拓扑优化解决了传统水平集方法数值计算复杂的问题,由于水平集拓扑优化需要引入更高级别的函数来构建拓扑模型,大部分关于参数化水平集拓扑优化的研究止于二维。文章在二维参数化水平集拓扑优化的基础上拓展到三维,结合点...参数化水平集拓扑优化解决了传统水平集方法数值计算复杂的问题,由于水平集拓扑优化需要引入更高级别的函数来构建拓扑模型,大部分关于参数化水平集拓扑优化的研究止于二维。文章在二维参数化水平集拓扑优化的基础上拓展到三维,结合点云的思想解决了三维拓扑构型的表达问题,并通过悬臂梁、简支梁等典型数值算例验证了算法能解决三维水平集无法引入更高维度函数等问题,亦讨论了引入SIMP(Solid Isotropic Material with Penalization)多材料插值模型的多相材料拓扑优化方法,以供多相材料拓扑优化方面研究进行参考。展开更多
针对传统故障模式和影响分析(failure mode and effect analysis,FMEA)方法存在评价使用精确数量化造成专家风险评估信息的丢失、忽略风险指标之间的相对重要性以及由于专家有限理性导致的评价固有的随机性等问题,利用区间值直觉模糊集...针对传统故障模式和影响分析(failure mode and effect analysis,FMEA)方法存在评价使用精确数量化造成专家风险评估信息的丢失、忽略风险指标之间的相对重要性以及由于专家有限理性导致的评价固有的随机性等问题,利用区间值直觉模糊集和云模型构建了一种改进的FMEA风险评估方法。首先,引入区间值直觉模糊集(IVIFS)来描述专家评价信息的复杂性和不确定性,通过运用区间值直觉模糊熵,计算专家权重和风险因子的权重;其次,采用云模型的方法,通过比较各支持云模型和反对云模型与正、负理想云模型的正、负相似度,获得故障模式评价值的综合相似度,通过对综合相似度大小排序得到各故障模式风险排序;最后,以自动扶梯的梯级、踏板和胶带风险评估为例进行分析,验证该评估方法的实用性和可行性。展开更多
文摘With the rapid development of the cloud computing technology, it has matured enough for a lot of individuals and organizations to move their work into the cloud. Correspondingly, a variety of cloud services are emerging. It is a key issue to assess the cloud services in order to help the cloud users select the most suitable cloud service and the cloud providers offer this service with the highest quality. The criteria parameters defining the cloud services are complex which lead to cloud service deviation. In this paper, we propose an assessment method of parameters importance in cloud services using rough set theory. The method can effectively compute the importance of cloud services parameters and sort them. On the one hand, the calculation can be used as the credible reference when users choose their appropriate cloud services. On the other hand, it can help cloud service providers to meet user requirements and enhance the user experience. The simulation results show the effectiveness of the method and its relevance in the cloud context.
文摘针对多目标识别过程中点云分类和分割精度不高的问题,提出了一种基于改进Transformer模型的点云分类与分割方法DRPT(Double randomness Point Transformer),该方法在Transformer模型卷积投影层创建新的点嵌入,利用局部邻域的动态处理在数据特征向量中持续增加全局特征属性,从而提高多目标识别中点云分类和分割的精度。实验中采用了标准基准数据集(ModelNet40、ShapeNet部分分割和SemanticKITTI场景语义分割数据集)以验证模型的性能,实验结果表明:DRPT模型的pIoU值为85.9%,比其他模型平均高出3.5%,有效提高了多目标识别检测时点云分类与分割精度,是对智能网联技术发展的有效支撑。
文摘参数化水平集拓扑优化解决了传统水平集方法数值计算复杂的问题,由于水平集拓扑优化需要引入更高级别的函数来构建拓扑模型,大部分关于参数化水平集拓扑优化的研究止于二维。文章在二维参数化水平集拓扑优化的基础上拓展到三维,结合点云的思想解决了三维拓扑构型的表达问题,并通过悬臂梁、简支梁等典型数值算例验证了算法能解决三维水平集无法引入更高维度函数等问题,亦讨论了引入SIMP(Solid Isotropic Material with Penalization)多材料插值模型的多相材料拓扑优化方法,以供多相材料拓扑优化方面研究进行参考。
文摘针对传统故障模式和影响分析(failure mode and effect analysis,FMEA)方法存在评价使用精确数量化造成专家风险评估信息的丢失、忽略风险指标之间的相对重要性以及由于专家有限理性导致的评价固有的随机性等问题,利用区间值直觉模糊集和云模型构建了一种改进的FMEA风险评估方法。首先,引入区间值直觉模糊集(IVIFS)来描述专家评价信息的复杂性和不确定性,通过运用区间值直觉模糊熵,计算专家权重和风险因子的权重;其次,采用云模型的方法,通过比较各支持云模型和反对云模型与正、负理想云模型的正、负相似度,获得故障模式评价值的综合相似度,通过对综合相似度大小排序得到各故障模式风险排序;最后,以自动扶梯的梯级、踏板和胶带风险评估为例进行分析,验证该评估方法的实用性和可行性。