In the formula of the current Mode,there is some blemish.The Mode does not present the concentrate trend of all the frequencies.It not only has no a consideration of the group interval,but also the further groups’ fr...In the formula of the current Mode,there is some blemish.The Mode does not present the concentrate trend of all the frequencies.It not only has no a consideration of the group interval,but also the further groups’ frequencies.In this paper,the author bring upped a new calculation formula of Mode.展开更多
针对卷积神经网络(Convolutional neural network,CNN)模型在对工业数值型数据分类方面存在特征使用不充分、模型分类性能不佳等问题,提出了一种基于自适应卷积核的改进CNN(Improved CNN based on adaptive convolution kernel, ACK-IC...针对卷积神经网络(Convolutional neural network,CNN)模型在对工业数值型数据分类方面存在特征使用不充分、模型分类性能不佳等问题,提出了一种基于自适应卷积核的改进CNN(Improved CNN based on adaptive convolution kernel, ACK-ICNN)算法。该算法为了增加特征的重复使用率,构建了一种多尺度卷积核的模型结构,通过融合处理卷积核提取的不同特征来实现,增强了模型的适应能力;为了进一步提升该算法的性能,利用网格搜索算法自适应选取CNN中最优的卷积核大小,使得模型能够提取出最优的特征。采用TE过程的故障数据对其进行测试,并与支持向量机、极限学习机、最近邻等典型的数据驱动方法进行对比,测试结果表明,该算法能有效提升各类故障的分类精度。展开更多
文摘In the formula of the current Mode,there is some blemish.The Mode does not present the concentrate trend of all the frequencies.It not only has no a consideration of the group interval,but also the further groups’ frequencies.In this paper,the author bring upped a new calculation formula of Mode.
文摘针对卷积神经网络(Convolutional neural network,CNN)模型在对工业数值型数据分类方面存在特征使用不充分、模型分类性能不佳等问题,提出了一种基于自适应卷积核的改进CNN(Improved CNN based on adaptive convolution kernel, ACK-ICNN)算法。该算法为了增加特征的重复使用率,构建了一种多尺度卷积核的模型结构,通过融合处理卷积核提取的不同特征来实现,增强了模型的适应能力;为了进一步提升该算法的性能,利用网格搜索算法自适应选取CNN中最优的卷积核大小,使得模型能够提取出最优的特征。采用TE过程的故障数据对其进行测试,并与支持向量机、极限学习机、最近邻等典型的数据驱动方法进行对比,测试结果表明,该算法能有效提升各类故障的分类精度。