It is revealed that the dynamic stability of 2-D recursive continuous-discrete systems with interval parameters involves the problem of robust Hurwitz-Schur stability of bivariate polynomials family. It is proved that...It is revealed that the dynamic stability of 2-D recursive continuous-discrete systems with interval parameters involves the problem of robust Hurwitz-Schur stability of bivariate polynomials family. It is proved that the Hurwitz-Schur stability of the denominator polynomials of the systems is necessary and sufficient for the asymptotic stability of the 2-D hybrid systems. The 2-D hybrid transformation, i. e. 2-D Laplace-Z transformation, has been proposed to solve the stability analysis of the 2-D continuous-discrete systems, to get the 2-D hybrid transfer functions of the systems. The edge test for the Hurwitz-Schur stability of interval bivariate polynomials is introduced. The Hurwitz-Schur stability of the interval family of 2-D polynomials can be guaranteed by the stability of its finite edge polynomials of the family. An algorithm about the stability test of edge polynomials is given.展开更多
深度学习(deep learning,DL)应用于推动智能故障诊断的发展,带来了显著的性能提升。然而,现有方法大多无法捕捉机械设备的时间信息和全局特征,无法收集到足够的故障信息。同时,由于运行环境复杂恶劣,单源故障诊断方法难以稳定、广泛地...深度学习(deep learning,DL)应用于推动智能故障诊断的发展,带来了显著的性能提升。然而,现有方法大多无法捕捉机械设备的时间信息和全局特征,无法收集到足够的故障信息。同时,由于运行环境复杂恶劣,单源故障诊断方法难以稳定、广泛地提取故障特征。因此,提出一种基于多源信息融合(multi-source information fusion,MSIF)的连续小波和TransXNet三相异步电机故障诊断新方法,通过提取和整合丰富的特征来提高诊断性能稳定性。首先,搭建三相异步电机故障实验平台,使用加速度传感器与电流传感器采集电机多种工况下的振动信号与电流信号,获得多源信号;其次,提出一种新的轻量级混合网络模块:双动态令牌混合器(dual dynamic token mixer,D-Mixer),其可以动态地利用全局和局部信息,同时注入大的感受野和强大的归纳偏差,而不牺牲输入依赖性,提高了特征提取能力。提出多尺度前馈网络(multi-scale feed-forward network,MS-FFN),在前馈网络中进行多尺度特征聚合。通过交替使用D-Mixer和MS-FFN,构建一种新型的混合CNN-Transformer网络:TransXNet;然后,利用连续小波变换将多源信号进行时频变换,提出数据级融合策略获得多源信息图,将多源信息图输入到TransXNet中进行特征分割以及聚合以完成特征提取,以训练并验证所提出的TransXNet有效性;最后,使用多源测试样本来验证所提出方法的诊断性能。结果表明,基于TransXNet强大的特征提取能力,识别准确率达到100%。通过对比调整兰德指数、归一化互信息、F1分数和准确率4个评价指标以及抗噪性分析,得出所提方法优于目前故障诊断领域最先进的方法(state of the art,SOTA),在故障诊断领域具有很好的前景。展开更多
基金This project was supported by National Natural Science Foundation of China (69971002).
文摘It is revealed that the dynamic stability of 2-D recursive continuous-discrete systems with interval parameters involves the problem of robust Hurwitz-Schur stability of bivariate polynomials family. It is proved that the Hurwitz-Schur stability of the denominator polynomials of the systems is necessary and sufficient for the asymptotic stability of the 2-D hybrid systems. The 2-D hybrid transformation, i. e. 2-D Laplace-Z transformation, has been proposed to solve the stability analysis of the 2-D continuous-discrete systems, to get the 2-D hybrid transfer functions of the systems. The edge test for the Hurwitz-Schur stability of interval bivariate polynomials is introduced. The Hurwitz-Schur stability of the interval family of 2-D polynomials can be guaranteed by the stability of its finite edge polynomials of the family. An algorithm about the stability test of edge polynomials is given.
文摘深度学习(deep learning,DL)应用于推动智能故障诊断的发展,带来了显著的性能提升。然而,现有方法大多无法捕捉机械设备的时间信息和全局特征,无法收集到足够的故障信息。同时,由于运行环境复杂恶劣,单源故障诊断方法难以稳定、广泛地提取故障特征。因此,提出一种基于多源信息融合(multi-source information fusion,MSIF)的连续小波和TransXNet三相异步电机故障诊断新方法,通过提取和整合丰富的特征来提高诊断性能稳定性。首先,搭建三相异步电机故障实验平台,使用加速度传感器与电流传感器采集电机多种工况下的振动信号与电流信号,获得多源信号;其次,提出一种新的轻量级混合网络模块:双动态令牌混合器(dual dynamic token mixer,D-Mixer),其可以动态地利用全局和局部信息,同时注入大的感受野和强大的归纳偏差,而不牺牲输入依赖性,提高了特征提取能力。提出多尺度前馈网络(multi-scale feed-forward network,MS-FFN),在前馈网络中进行多尺度特征聚合。通过交替使用D-Mixer和MS-FFN,构建一种新型的混合CNN-Transformer网络:TransXNet;然后,利用连续小波变换将多源信号进行时频变换,提出数据级融合策略获得多源信息图,将多源信息图输入到TransXNet中进行特征分割以及聚合以完成特征提取,以训练并验证所提出的TransXNet有效性;最后,使用多源测试样本来验证所提出方法的诊断性能。结果表明,基于TransXNet强大的特征提取能力,识别准确率达到100%。通过对比调整兰德指数、归一化互信息、F1分数和准确率4个评价指标以及抗噪性分析,得出所提方法优于目前故障诊断领域最先进的方法(state of the art,SOTA),在故障诊断领域具有很好的前景。
