This paper proposes novel multi-layer neural networks based on Independent Component Analysis for feature extraction of fault modes. By the use of ICA, invariable features embedded in multi-channel vibration measureme...This paper proposes novel multi-layer neural networks based on Independent Component Analysis for feature extraction of fault modes. By the use of ICA, invariable features embedded in multi-channel vibration measurements under different operating conditions (rotating speed and/or load) can be captured together.Thus, stable MLP classifiers insensitive to the variation of operation conditions are constructed. The successful results achieved by selected experiments indicate great potential of ICA in health condition monitoring of rotating machines.展开更多
Citric acid is an important organic substance whose marketing concerns various fields. Nevertheless, until 1997 the scientific literature reported little information about the process of crystallization by cooling thr...Citric acid is an important organic substance whose marketing concerns various fields. Nevertheless, until 1997 the scientific literature reported little information about the process of crystallization by cooling through which the commercial product is obtained. In particular, the available studies were aimed to investigate only the kinetics of nucleation and crystal growth neglecting some effective aspects of the industrial crystallization in mechanically stirred apparatus. In order to fill that sci-tech gap, the Department of Chemical Engineering at the University ＂La Sapienza＂ of Rome decided to lead a long and meticulous experimental research on the crystallization in discontinuous （batch） of CAM （citric acid monohydrate） in the allotropic form that is stable at room temperature. Due to the number of people involved in that pioneering work, carried out in the historic laboratories of＂La Sapienza＂ （Faculty of Engineering）, and motivated by the publication of related M.Sc. dissertations and research papers, such collective effort was called ＂School of Industrial Crystallization＂. Among the graduate students in Chemical Engineering that 17 years ago participated in that fruitful experience there was also the author who, under the supervision of Prof. Barbara Mazzarotta, had the specific task of assessing the effects on CAM of changing the crystallization operating conditions until their optimization; the achievements are briefly illustrated in this paper.展开更多
The kinetics is analyzed of the drift of non-potential plasma waves in spatial positions and wavevectors due to plasma's spatial inhomogeneity. The analysis is based on highly informative kinetic scenarios of the ...The kinetics is analyzed of the drift of non-potential plasma waves in spatial positions and wavevectors due to plasma's spatial inhomogeneity. The analysis is based on highly informative kinetic scenarios of the drift of electromagnetic waves in a cold ionized plasma in the absence of a magnetic field(Erofeev 2015 Phys. Plasmas 22 092302) and the drift of long Langmuir waves in a cold magnetized plasma(Erofeev 2019 J. Plasma Phys. 85 905850104). It is shown that the traditional concept of the wave kinetic equation does not account for the effects of the forced plasma oscillations that are excited when the waves propagate in an inhomogeneous plasma.Terms are highlighted that account for these oscillations in the kinetic equations of the abovementioned highly informative wave drift scenarios.展开更多
针对长时间序列电力负荷的预测精度低的问题,应用了基于Informer长时间序列模型的电力负荷预测方法.该方法通过Informer模型中的自注意力蒸馏机制,使得每层的解码器都将输入序列的长度缩短一半,从而极大地节约了Encoder内存开销,并在编...针对长时间序列电力负荷的预测精度低的问题,应用了基于Informer长时间序列模型的电力负荷预测方法.该方法通过Informer模型中的自注意力蒸馏机制,使得每层的解码器都将输入序列的长度缩短一半,从而极大地节约了Encoder内存开销,并在编码器结构中使用生成式结构,使得预测解码时间极大的缩短;以澳大利亚的电力负荷数据作为测试用例,并与长短时记忆神经网络(long-short term memory,LSTM)和卷积神经网络(convolutional neural network,CNN)模型方法进行对比,结果表明,Informer模型的预测精度更高,Pearson相关系数可以达到91.30%,有效提高了负荷预测精度.展开更多
文摘This paper proposes novel multi-layer neural networks based on Independent Component Analysis for feature extraction of fault modes. By the use of ICA, invariable features embedded in multi-channel vibration measurements under different operating conditions (rotating speed and/or load) can be captured together.Thus, stable MLP classifiers insensitive to the variation of operation conditions are constructed. The successful results achieved by selected experiments indicate great potential of ICA in health condition monitoring of rotating machines.
文摘Citric acid is an important organic substance whose marketing concerns various fields. Nevertheless, until 1997 the scientific literature reported little information about the process of crystallization by cooling through which the commercial product is obtained. In particular, the available studies were aimed to investigate only the kinetics of nucleation and crystal growth neglecting some effective aspects of the industrial crystallization in mechanically stirred apparatus. In order to fill that sci-tech gap, the Department of Chemical Engineering at the University ＂La Sapienza＂ of Rome decided to lead a long and meticulous experimental research on the crystallization in discontinuous （batch） of CAM （citric acid monohydrate） in the allotropic form that is stable at room temperature. Due to the number of people involved in that pioneering work, carried out in the historic laboratories of＂La Sapienza＂ （Faculty of Engineering）, and motivated by the publication of related M.Sc. dissertations and research papers, such collective effort was called ＂School of Industrial Crystallization＂. Among the graduate students in Chemical Engineering that 17 years ago participated in that fruitful experience there was also the author who, under the supervision of Prof. Barbara Mazzarotta, had the specific task of assessing the effects on CAM of changing the crystallization operating conditions until their optimization; the achievements are briefly illustrated in this paper.
文摘The kinetics is analyzed of the drift of non-potential plasma waves in spatial positions and wavevectors due to plasma's spatial inhomogeneity. The analysis is based on highly informative kinetic scenarios of the drift of electromagnetic waves in a cold ionized plasma in the absence of a magnetic field(Erofeev 2015 Phys. Plasmas 22 092302) and the drift of long Langmuir waves in a cold magnetized plasma(Erofeev 2019 J. Plasma Phys. 85 905850104). It is shown that the traditional concept of the wave kinetic equation does not account for the effects of the forced plasma oscillations that are excited when the waves propagate in an inhomogeneous plasma.Terms are highlighted that account for these oscillations in the kinetic equations of the abovementioned highly informative wave drift scenarios.
文摘针对长时间序列电力负荷的预测精度低的问题,应用了基于Informer长时间序列模型的电力负荷预测方法.该方法通过Informer模型中的自注意力蒸馏机制,使得每层的解码器都将输入序列的长度缩短一半,从而极大地节约了Encoder内存开销,并在编码器结构中使用生成式结构,使得预测解码时间极大的缩短;以澳大利亚的电力负荷数据作为测试用例,并与长短时记忆神经网络(long-short term memory,LSTM)和卷积神经网络(convolutional neural network,CNN)模型方法进行对比,结果表明,Informer模型的预测精度更高,Pearson相关系数可以达到91.30%,有效提高了负荷预测精度.
