In many batch processes, there are related or independence relationships among process variables. The traditional monitoring method usually carries out a single statistical model according to the related or independen...In many batch processes, there are related or independence relationships among process variables. The traditional monitoring method usually carries out a single statistical model according to the related or independent method, and in the feature extraction there is not fully taken into account the characterization of fault information, it will make the process monitoring ineffective, so a fault monitoring method based on WGNPE(weighted global neighborhood preserving embedding)–GSVDD(greedy support vector data description) related and independent variables is proposed. First, mutual information method is used to separate the related variables and independent variables. Secondly, WGNPE method is used to extract the local and global structures of the related variables in batch process and highlight the fault information, GSVDD method is used to extract the process information of the independent variables quickly and effectively. Finally, the statistical monitoring model is established to achieve process monitoring based on WGNPE and GSVDD. The effectiveness of the proposed method was verified by the penicillin fermentation process.展开更多
In this paper,a case study is carried out in comparison of pipes-and-filters architecture and batch sequential architecture.Concepts on a data flow system and the two mentioned architectures are presented.A Java templ...In this paper,a case study is carried out in comparison of pipes-and-filters architecture and batch sequential architecture.Concepts on a data flow system and the two mentioned architectures are presented.A Java template class design in implementing the "pipes" and "filters" in the pipes-and-filters architecture is given at the design level.Finally,this paper uses a concrete example to show how to use Java to implement the pipesand-filters architecture.Using varied amount of data from text files,performance and memory usage of the two architectures are illustrated.展开更多
为解决传统神经网络在CIFAR-10(Canadian Institute For Advanced Research)数据集上进行图像分类识别时,存在的模型准确率较低和训练过程易发生过拟合现象等问题,提出了一种将卷积神经网络和批归一化相结合的新神经网络结构构建方法。...为解决传统神经网络在CIFAR-10(Canadian Institute For Advanced Research)数据集上进行图像分类识别时,存在的模型准确率较低和训练过程易发生过拟合现象等问题,提出了一种将卷积神经网络和批归一化相结合的新神经网络结构构建方法。该方法首先对数据集进行数据增强和边界填充处理,其次对典型的CNN(Convolutional Neural Networks)网络结构进行改进,移除了卷积层组中的池化层,仅保留了卷积层和BN(Batch Normalization)层,并适量增加卷积层组。为了验证模型的有效性和准确性,设计了6组不同的神经网络结构对模型进行训练。实验结果表明,在相同训练周期数下,推荐使用的model-6模型表现最佳,测试准确率高达90.17%,突破了长期以来经典CNN在CIFAR-10数据集上难于达到90%准确率的瓶颈,为图像分类识别提供了新的解决方案和模型参考。展开更多
基金Supported by the National Natural Science Foundation of China(No.61763029)the Natural Science Foundation of Gansu Province(1610RJZA016)
文摘In many batch processes, there are related or independence relationships among process variables. The traditional monitoring method usually carries out a single statistical model according to the related or independent method, and in the feature extraction there is not fully taken into account the characterization of fault information, it will make the process monitoring ineffective, so a fault monitoring method based on WGNPE(weighted global neighborhood preserving embedding)–GSVDD(greedy support vector data description) related and independent variables is proposed. First, mutual information method is used to separate the related variables and independent variables. Secondly, WGNPE method is used to extract the local and global structures of the related variables in batch process and highlight the fault information, GSVDD method is used to extract the process information of the independent variables quickly and effectively. Finally, the statistical monitoring model is established to achieve process monitoring based on WGNPE and GSVDD. The effectiveness of the proposed method was verified by the penicillin fermentation process.
文摘In this paper,a case study is carried out in comparison of pipes-and-filters architecture and batch sequential architecture.Concepts on a data flow system and the two mentioned architectures are presented.A Java template class design in implementing the "pipes" and "filters" in the pipes-and-filters architecture is given at the design level.Finally,this paper uses a concrete example to show how to use Java to implement the pipesand-filters architecture.Using varied amount of data from text files,performance and memory usage of the two architectures are illustrated.
文摘为解决传统神经网络在CIFAR-10(Canadian Institute For Advanced Research)数据集上进行图像分类识别时,存在的模型准确率较低和训练过程易发生过拟合现象等问题,提出了一种将卷积神经网络和批归一化相结合的新神经网络结构构建方法。该方法首先对数据集进行数据增强和边界填充处理,其次对典型的CNN(Convolutional Neural Networks)网络结构进行改进,移除了卷积层组中的池化层,仅保留了卷积层和BN(Batch Normalization)层,并适量增加卷积层组。为了验证模型的有效性和准确性,设计了6组不同的神经网络结构对模型进行训练。实验结果表明,在相同训练周期数下,推荐使用的model-6模型表现最佳,测试准确率高达90.17%,突破了长期以来经典CNN在CIFAR-10数据集上难于达到90%准确率的瓶颈,为图像分类识别提供了新的解决方案和模型参考。
