Sun Yaqiang

TL;DR: A novel eigenvector based on time–frequency-wavelet joint features is proposed to effectively represent bearing degradation process and a deep autoencoder based joint features compression and computing method is presented to retain effective information without increasing the scale of DNN.

TL;DR: An integrated deep learning approach for multi-bearing remaining useful life collaborative prediction by combining both time domain features and frequency domain features is proposed, which can extract high-quality degradation patterns of rolling bearing from vibration signals.

TL;DR: A new method for the prediction of bearing RUL based on deep convolution neural network (CNN) with a new feature extraction method to obtain the eigenvector, named the spectrum-principal-energy-vector, which is suitable for deep CNN.

TL;DR: In this paper, a bearing feature extraction method and a method and device for predicting the remaining service life of a bearing were presented, and the method comprises the steps: obtaining a frequency domain sequence of the bearing according to a time domain vibration signal of bearing, and carrying out the principal energy compression of the high-dimension feature data in the frequency-domain sequence through a code compression model corresponding to the frequencydomain sequence to obtain low-dimensional feature data; extracting a part playing a leading role in frequency-domains sequence, and shortening the data structure to achieve the data optimization

TL;DR: In this article, a method and apparatus for predicting a bus passenger flow is presented, which comprises: passenger card-swiping information of a to-be-predicted route is obtained, wherein the passenger identifier information and card swiping time are determined, according to which the total bus taking number of times of each passenger is determined, and passengers are classified into regular passengers, medium passengers, and random passengers.