Northeastern University (China)

About: Northeastern University (China) is a education organization based out in Shenyang, China. It is known for research contribution in the topics: Control theory & Microstructure. The organization has 36087 authors who have published 36125 publications receiving 426807 citations. The organization is also known as: Dōngběi Dàxué & Northeastern University (东北大学).

Resilience and Friability of Transportation Networks: Evaluation, Analysis and Optimization

Abstract: To analyze the resilience of transportation networks, it is proposed to use a quantifiable resilience evaluation approach. First, we represent transportation networks by an undirected graph with the nodes as cities and edges as traffic roads. Because the survivability of transportation of a pair of cities depends on the number of passageways between them, the resilience of a city node can be evaluated by the weighted average number of reliable passageways with all other city nodes in the network. The network resilience can then be calculated by the weighted sum of the resilience of all nodes. To identify critical road lines or hub cities in networks, the concept of friability is proposed. This is defined as the reduction in total resilience upon removing an edge or hub city. Following the resilience and friability evaluation, a structure optimization model with a computational algorithm for transportation network design is recommended. Based on the recommended approaches, the resilience and friability of the railway network within the Chinese mainland is evaluated and analyzed. Several interesting conclusions are drawn from the computational results. The friability value of the railway lines in the Sichuan Basin which was damaged by the recent earthquake in China was also calculated.

Diagnosing Rotating Machines With Weakly Supervised Data Using Deep Transfer Learning

Abstract: Rotating machinery fault diagnosis problems have been well-addressed when sufficient supervised data of the tested machine are available using the latest data-driven methods. However, it is still challenging to develop effective diagnostic method with insufficient training data, which is highly demanded in real-industrial scenarios, since high-quality data are usually difficult and expensive to collect. Considering the underlying similarities of rotating machines, data mining on different but related equipments potentially benefit the diagnostic performance on the target machine. Therefore, a novel transfer learning method for diagnostics based on deep learning is proposed in this article, where the diagnostic knowledge learned from sufficient supervised data of multiple rotating machines is transferred to the target equipment with domain adversarial training. Different from the existing studies, a more generalized transfer learning problem with different label spaces of domains is investigated, and different fault severities are also considered in fault diagnostics. The experimental results on four datasets validate the effectiveness of the proposed method, and show it is feasible and promising to explore different datasets to improve diagnostic performance.