Death is one of the urgent crises event in human society that occur during the lifecycle of each individual. It has anintegral relation with religion, especially with rites and rituals, through which the deceased pers...Death is one of the urgent crises event in human society that occur during the lifecycle of each individual. It has anintegral relation with religion, especially with rites and rituals, through which the deceased person is appeased withthe intervention of supernatural. FurtherInore, the death rites, popularly known as funeral rites, which incorporatethe deceased into the world of the dead are more extensively elaborated and assigned the greatest importance.Mourning is integral element related with the death and during the event social life is suspended for all thoseaffected by it and length of the period increases with the closeness of social ties with the deceased. In every society,there are certain customs related to death, as well as disposal of the corpse which reflect the parochial belief systemassociated with the event. In this paper, an attempt has been made to evaluate the customs associated with disposalof the death, integral parochial religious rites and rituals among the Mishings of Upper Assam, India.展开更多
为了实现轮毂焊缝缺陷的智能化检测,本文对深度学习目标检测算法(You Only Look Once version3,YOLOv3)进行改进,得到YOLOv3-MC算法用于轮毂焊缝缺陷的检测。首先,使用工业相机采集轮毂焊缝图像,然后标注图像制作数据集,并且通过数据增...为了实现轮毂焊缝缺陷的智能化检测,本文对深度学习目标检测算法(You Only Look Once version3,YOLOv3)进行改进,得到YOLOv3-MC算法用于轮毂焊缝缺陷的检测。首先,使用工业相机采集轮毂焊缝图像,然后标注图像制作数据集,并且通过数据增强方法扩充数据集。接着,为了提高算法检测精度,使用Mish激活函数替换YOLOv3主干网络中的激活函数。修改算法的损失函数,使用完备交并比(Complete Intersection over Union,CIoU)的计算方法提升算法检测的定位精度。最后使用训练集训练算法模型,再使用验证集和测试集图像数据进行检测试验,结果表明,YOLOv3-MC的最优模型在验证集上的平均准确率(Mean Average Precision,mAP)达到了98.94%,F1得分值为0.99,平均交并比(Average Intersection over Union,AvgIoU)为80.92%,检测速度为76.59帧/秒,模型大小234MB。该模型在测试集上的检测正确率达到了99.29%。相较于传统机器视觉检测方法,该方法提高了检测精度,满足轮毂生产企业的焊缝实时在线检测需求。展开更多
针对现有模型仅考虑一种内部状态对锂电池性能退化影响的问题,同步建立3个模型分别预测3种时变状态随锂电池性能退化的变化轨迹,并以内阻与温度预测为基础实现锂电池容量的实时更新;针对传统神经网络中的sigmoid与ReLU激活函数存在梯度...针对现有模型仅考虑一种内部状态对锂电池性能退化影响的问题,同步建立3个模型分别预测3种时变状态随锂电池性能退化的变化轨迹,并以内阻与温度预测为基础实现锂电池容量的实时更新;针对传统神经网络中的sigmoid与ReLU激活函数存在梯度消失与神经元坏死问题,在双向长短时记忆(bi-directional long short term memory,Bi-LSTM)网络与全连接网络中引入一种新的Mish激活函数,使模型以平稳的梯度流提取更高质量的特征用于剩余使用寿命(RUL)预测的建模分析。最后利用蒙特卡洛(Monte Carlo,MC)与Dropout技术对锂电池RUL的预测结果不确定性进行分析。在美国Kristen教授课题组所公开的锂电池数据集上进行对比试验的结果表明,所提改进Bi-LSTM模型预测的均方误差(mean squared error,MSE)、平均绝对误差(mean absolute error,MAE)与R^(2)可达9.16×10^(-5)、0.00795、99.794%。随着获取数据量的增加,模型对锂电池RUL预测的精度越高,RUL平均预测误差可达2.3个循环,验证了所提模型能有效地实现锂电池循环RUL的实时更新。展开更多
文摘Death is one of the urgent crises event in human society that occur during the lifecycle of each individual. It has anintegral relation with religion, especially with rites and rituals, through which the deceased person is appeased withthe intervention of supernatural. FurtherInore, the death rites, popularly known as funeral rites, which incorporatethe deceased into the world of the dead are more extensively elaborated and assigned the greatest importance.Mourning is integral element related with the death and during the event social life is suspended for all thoseaffected by it and length of the period increases with the closeness of social ties with the deceased. In every society,there are certain customs related to death, as well as disposal of the corpse which reflect the parochial belief systemassociated with the event. In this paper, an attempt has been made to evaluate the customs associated with disposalof the death, integral parochial religious rites and rituals among the Mishings of Upper Assam, India.
文摘为了实现轮毂焊缝缺陷的智能化检测,本文对深度学习目标检测算法(You Only Look Once version3,YOLOv3)进行改进,得到YOLOv3-MC算法用于轮毂焊缝缺陷的检测。首先,使用工业相机采集轮毂焊缝图像,然后标注图像制作数据集,并且通过数据增强方法扩充数据集。接着,为了提高算法检测精度,使用Mish激活函数替换YOLOv3主干网络中的激活函数。修改算法的损失函数,使用完备交并比(Complete Intersection over Union,CIoU)的计算方法提升算法检测的定位精度。最后使用训练集训练算法模型,再使用验证集和测试集图像数据进行检测试验,结果表明,YOLOv3-MC的最优模型在验证集上的平均准确率(Mean Average Precision,mAP)达到了98.94%,F1得分值为0.99,平均交并比(Average Intersection over Union,AvgIoU)为80.92%,检测速度为76.59帧/秒,模型大小234MB。该模型在测试集上的检测正确率达到了99.29%。相较于传统机器视觉检测方法,该方法提高了检测精度,满足轮毂生产企业的焊缝实时在线检测需求。
文摘针对现有模型仅考虑一种内部状态对锂电池性能退化影响的问题,同步建立3个模型分别预测3种时变状态随锂电池性能退化的变化轨迹,并以内阻与温度预测为基础实现锂电池容量的实时更新;针对传统神经网络中的sigmoid与ReLU激活函数存在梯度消失与神经元坏死问题,在双向长短时记忆(bi-directional long short term memory,Bi-LSTM)网络与全连接网络中引入一种新的Mish激活函数,使模型以平稳的梯度流提取更高质量的特征用于剩余使用寿命(RUL)预测的建模分析。最后利用蒙特卡洛(Monte Carlo,MC)与Dropout技术对锂电池RUL的预测结果不确定性进行分析。在美国Kristen教授课题组所公开的锂电池数据集上进行对比试验的结果表明,所提改进Bi-LSTM模型预测的均方误差(mean squared error,MSE)、平均绝对误差(mean absolute error,MAE)与R^(2)可达9.16×10^(-5)、0.00795、99.794%。随着获取数据量的增加,模型对锂电池RUL预测的精度越高,RUL平均预测误差可达2.3个循环,验证了所提模型能有效地实现锂电池循环RUL的实时更新。
