Capital Normal University

About: Capital Normal University is a education organization based out in Beijing, China. It is known for research contribution in the topics: Terahertz radiation & Quantum entanglement. The organization has 11441 authors who have published 11988 publications receiving 159071 citations. The organization is also known as: Shǒudū Shīfàn Dàxué.

Naive Bayes Bearing Fault Diagnosis Based on Enhanced Independence of Data

Abstract: The bearing is the key component of rotating machinery, and its performance directly determines the reliability and safety of the system. Data-based bearing fault diagnosis has become a research hotspot. Naive Bayes (NB), which is based on independent presumption, is widely used in fault diagnosis. However, the bearing data are not completely independent, which reduces the performance of NB algorithms. In order to solve this problem, we propose a NB bearing fault diagnosis method based on enhanced independence of data. The method deals with data vector from two aspects: the attribute feature and the sample dimension. After processing, the classification limitation of NB is reduced by the independence hypothesis. First, we extract the statistical characteristics of the original signal of the bearings effectively. Then, the Decision Tree algorithm is used to select the important features of the time domain signal, and the low correlation features is selected. Next, the Selective Support Vector Machine (SSVM) is used to prune the dimension data and remove redundant vectors. Finally, we use NB to diagnose the fault with the low correlation data. The experimental results show that the independent enhancement of data is effective for bearing fault diagnosis.

Spiky and transition layer steady states of chemotaxis systems via global bifurcation and Helly's compactness theorem.

Abstract: The most important phenomenon in chemotaxis is cell aggregation. To model this phenomenon we use spiky or transition layer (step-function-like) steady states. In the case of one spatial dimension, we carry out global bifurcation analysis on the Keller-Segel model and several variants of it, showing that positive steady states exist if the chemotactic coefficient χ is larger than a bifurcation value χ1 which can be explicitly expressed in terms of the parameters in the models; then we use Helly's compactness theorem to obtain the profiles of these steady states when the ratio of the chemotactic coefficient and the cell diffusion rate is large, showing that they are either spiky or have the transition layer structure. Our results provide insights on how the biological parameters affect pattern formation, and reveal the similarities and differences of some popular chemotaxis models.

Induction of sporulation in plant pathogenic fungi

Abstract: Spore morphologies are a major character in fungal taxonomy, although many isolates are not able to sporulate on common artificial media. This article reviews the effect of nutrition, host tissue, and light on fungal sporulation in artificial media. A trial experiment using 42 strains that failed to sporulate on potato dextrose agar (PDA) and half-strength PDA after 3 months is reported. Five strategies (1/10-strength PDA, CaCO3 water agar, pine needle medium, mulberry agar, and near-ultraviolet light irradiation) were applied to induce these strains to sporulate, with an overall success rate of 62%. Pine needle medium was the most successful method, which induced sporulation of 40% of recalcitrant strains.

Deep transfer learning for military object recognition under small training set condition

Abstract: Convolutional neural network is powerful for general object recognition. However, its excellent performance depends largely on huge training set. Facing task like military object recognition in which image samples for training are scarce, its performance will degrade sharply. To solve this problem, a deep transfer learning method is proposed in this paper. The main idea consists of two parts: transfer learning for prior knowledge embedding and mixed layer for better feature extraction. It has been proved that the ability of feature extraction learned in large dataset is helpful to related tasks and can be transferred to a new neural network. The transfer learning process is achieved by fixing the weights of some layers and then retraining the remained layers. The key problem for deep transfer learning is which part should be transferred and which part should be retrained to adapt the network to the new task. This problem is solved by extensive experiments, and it is found that retraining the last three layers and transferring prior to the other layers can reach the best performance. Besides, we used mixed layer scheme to make use of the current information. In each mixed layer, convolution filters in different scales are combined together, helping to adapt features in different scales. By employing these two methods, the proposed method exhibits a large improvement in military object recognition under small training set. Experiments demonstrate that our method can achieve a high recognition precision, superior to many other algorithms compared.