Search results

The current studies on remaining useful life (RUL) prediction mainly rely on convolutional neural networks (CNNs) and long short-term memories (LSTMs) and do not take full advantage of the attention mechanism, resulting in lack of prediction accuracy. To further improve the performance of the above models, this study aims to propose a novel end-to-end RUL prediction framework, called convolutional recurrent attention network (CRAN) to achieve high accuracy.

The proposed CRAN is a CNN-LSTM-based model that effectively combines the powerful feature extraction ability of CNN and sequential processing capability of LSTM. The channel attention mechanism, spatial attention mechanism and LSTM attention mechanism are incorporated in CRAN, assigning different attention coefficients to CNN and LSTM. First, features of the bearing vibration data are extracted from both time and frequency domain. Next, the training and testing set are constructed. Then, the CRAN is trained offline using the training set. Finally, online RUL estimation is performed by applying data from the testing set to the trained CRAN.

CNN-LSTM-based models have higher RUL prediction accuracy than CNN-based and LSTM-based models. Using a combination of max pooling and average pooling can reduce the loss of feature information, and in addition, the structure of the serial attention mechanism is superior to the parallel attention structure. Comparing the proposed CRAN with six different state-of-the-art methods, for the predicted results of two testing bearings, the proposed CRAN has an average reduction in the root mean square error of 57.07/80.25%, an average reduction in the mean absolute error of 62.27/85.87% and an average improvement in score of 12.65/6.57%.

This article provides a novel end-to-end rolling bearing RUL prediction framework, which can provide a reference for the formulation of bearing maintenance programs in the industry.

Does the choice of listing process matter in determining a firm's future crash risk? It is understood that the main function of an equity market is to provide price discovery, however, it is not clear whether the choice of listing methods would matter to the shareholders' wealth in the long term. We are the first to answer this question by utilising a hand-collected dataset that identifies all companies that went public via reverse merger (RM) in a growing emerging market.

Using hand-collected data from 2000 to 2018 in China, we follow the literature to construct two crash risk measures for RM and IPO firms. Our main analysis is performed using OLS regressions on the full sample as well as a sample using Propensity Score Matching. Our results hold with a number of robustness checks.

We find that reverse merger (RM) firms exhibit higher future stock price crash risk than initial public offering (IPO) firms. This relationship is more predominant in non-state-owned enterprises, and we find weak evidence suggesting such relationship weakens as firms stay longer in the market. There is no significant difference in future stock price crash risk between RM firms listed during IPO suspension periods and normal IPO firms.

We are the first to study the choice of listing method and its impact on firms' future stock price crash risk.

Cobalt is an extremely useful element, but there are very few separate cobalt deposits in China and imported cobalt ores are usually toxic. In order to develop more low-toxicity cobalt resources, China has to encourage exploration and development of this element. Samples are taken from Hanxing iron ore tailings and analyzed by ICP-MS and barium chloride titration experiments. The results indicate that the cobalt content is relatively high in Hanxing iron ore tailings, and some exceed the industrial grade. Therefore, with the depletion of cobalt resources, iron ore tailings are bound to become an important resource in China, and as these tailings are abundant in the Hanxing area, this area is expected to be of high development value.