Search

In the prediction of rolling bearing degradation trend, traditional feature selection mainly relies on manual experience and a single evaluation algorithm, which is easy to cause under-selection or misselection of features. Moreover, a single deep learning network cannot fully mine the performance degradation information contained in the data, which results in low prediction accuracy of the model. To solve the above problem, a prediction method of rolling bearing degradation trend based on adaptive Mahalanobis space (AMS) and fusion deep learning network is proposed. Firstly, the original signals are decomposed and the correlated kurtosis coefficient criterion is used to screen the intrinsic mode function (IMF) to reconstruct the new signals, and the features are extracted from the multi-domain perspective. Secondly, the multi-objective feature selection algorithm based on AMS is built to optimize characteristic automatically. With the aim of reducing manual dependencies, and serengthening the adaptability and generalization, Mahalanobis distance (MD) is combined with the exponential weighted moving average (EWMA) method to well characterize the degradation trend of bearing performance. Finally, the fusion model of sparse auto encoder and gated recurrent unit (SAE-GRU) is used for prediction. The experimental results show that the proposed method can effectively screen the optimal features and significantly improve the prediction accuracy. [ABSTRACT FROM AUTHOR]

Infrared hyperspectral is susceptible to clouds, and accurately identifying whether the hyperspectral infrared sounder data is polluted by clouds is of great significance for numerical weather prediction and atmospheric parameter inversion. Since the complex spectral characteristics of clouds, the existing spectral threshold methods and machine learning methods have the difficulties of undetermined threshold and clear field of view (FOV) missed and false detections. In order to improve the cloud recognition accuracy of infrared hyperspectral data, three end-to-end cloud detection models combining deep neural network (DNN) and convolutional neural network (CNN) and long short-term memory network (LSTM) are proposed. In this paper, taking the High Spectral Infrared Atmospheric Sounder (HIRAS) equipped with Fengyun-3D (FY-3D) satellite as the research object, based on the same platform Moderate Resolution Spectral Imager-II (MERSI-II) cloud mask (CLM) product, the HIRAS Cloud dataset is established, and the accuracy test and qualitative analysis are carried out by using the test datasets and Typhoon Siamba, July 3, 2022, as well as the earth observation scene under the conditions of ice and snow surface. The test datasets analysis results show that the cloud detection accuracy of CNN and CNN-LSTM model is stable at 0.96, and the false alarm rate of cloud is 0.035 and 0.036, respectively, and the detection ability of DNN model is slightly inferior to the former two in the same hidden layer, with an accuracyof 0.94. In further qualitative research, we found that the CNN-LSTM model has high accuracy and robustness in infrared hyperspectral cloud detection, and the detection results in a variety of surface scenarios are consistent with the actual situation of whether clouds occur in the FOV of the instrument. Compared with CLM products, it can better identify clear ocean scenes, and provide fast and efficient cloud detection reference for data assimilation systems. [ABSTRACT FROM AUTHOR]