Empirical wavelet transform (EWT) is a novel adaptive signal decomposition method, whose main shortcoming is the fact that Fourier segmentation is strongly dependent on the local maxima of the amplitudes of the Fourier spectrum. An enhanced empirical wavelet transform (MSCEWT) based on maximum-minimum length curve method is proposed to realize fault diagnosis of motor bearings. The maximum-minimum length curve method transforms the original vibration signal spectrum to scale space in order to obtain a set of minimum length curves, and find the maximum length curve value in the set of the minimum length curve values for obtaining the number of the spectrum decomposition intervals. The MSCEWT method is used to decompose the vibration signal into a series of intrinsic mode functions (IMFs), which are processed by Hilbert transform. Then the frequency of each component is extracted by power spectrum and compared with the theoretical value of motor bearing fault feature frequency in order to determine and obtain fault diagnosis result. In order to verify the effectiveness of the MSCEWT method for fault diagnosis, the actual motor bearing vibration signals are selected and the empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) methods are selected for comparative analysis in here. The results show that the maximum-minimum length curve method can enhance EWT method and the MSCEWT method can solve the shortcomings of the Fourier spectrum segmentation and can effectively decompose the bearing vibration signal for obtaining less number of intrinsic mode function (IMF) components than the EMD and EEMD methods. It can effectively extract the fault feature frequency of the motor bearing and realize fault diagnosis. Therefore, the study provides a new method for fault diagnosis of rotating machinery.

Empirical Wavelet Transform (EWT) is a novel non-stationary signal analysis method that can effectively identify different mode components in signals. However, due to the lack of processing noise and unstable signals caused by the Fourier spectrum adaptive segmentation problem, an improved EWT (FCMEWT) method based on the scale space threshold method and fuzzy C-means is proposed to decompose the vibration signal into an empirical mode with physical meaning. The FCMEWT method firstly scales the spectrum of the original vibration signal, and then uses the fuzzy C-means method to classify the spectrum in order to obtain the spectrum division interval. The vibration signal is decomposed into a set of intrinsic mode functions (IMFs) components, which are performed Hilbert transform for extracting the frequency of each component through the power spectrum. Finally, Pearson correlation coefficient between each IMF component and the original signal is calculated to obtain the correlation coefficient threshold in order to determine the final IMF component. In order to verify the effectiveness of FCMEWT method, the vibration signal motor bearing is selected in this paper. The FCMEWT method is compared with the empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) methods. The results show that the FCMEWT method can effectively solve the problem of Fourier spectrum segmentation in the EWT method, takes on better adaptive segmentation characteristics, and can effectively extract fault feature frequency of motor bearing. The fault diagnosis method can not only effectively extract motor bearing fault characteristics, but also has better diagnosis result than EMD and EEMD methods.

No abstract

Intelligent marketing and recommendation are a core business of commercial companies, and accurate prediction of sales is the premise and foundation for greater efficiency of smart marketing and recommendation. In order to predict product sales, deep neural network (DNN), convolutional neural network (CNN), time series analysis and other methods have been put forward, but most of which only focus on the temporal or spatial characteristics of data. According to modeling and analyzing sales of products, they are closely related to the spatial location and time of the corresponding merchants. The goal is to predict the sales of products accurately at a given time and place, we advance a hybrid model of CNN-LSTM to forecast sales. Firstly, a large-scale knowledge graph system based on merchants is constructed, which describes the sales data and the relevant interaction scenarios of the corresponding business, merchants and users through the data model of a graph, and add the spatial and data characteristics of the business data on the graph model to describe the temporal and spatial characteristics of the merchants. Based on the constructed business knowledge graph, graph convolutional neural network (GCN) is used to aggregate information and obtain spatial features. Correspondingly, long short-term memory (LSTM) is used to extract time features. Researchers combine the two characteristics to make the sales forecast. In this study, neural network and GCN-LSTM algorithm are respectively used to carry out experiments on two kinds of product regulations. The result shows that the sales predicted by hybrid model of GCN-LSTM is almost as equal as the actual sales. The average accuracy of the proposed model is 89%.