汇报时间：2019年9月12日（星期四）20：00
汇报地点：曲江校区西五楼南205会议室
汇报人：焦金阳、丁传仓、梁凯旋
国际会议信息
会议名称：32nd International Congress and Exhibition on Condition Monitoring and Diagnostic Engineering Management
会议时间：2-5 September，2019
会议地点：Huddersfield in West Yorkshire, United Kingdom
会议简介：COMADEM International has been successfully organising annual congresses and exhibitions in the UK and worldwide for 32 years. COMADEM International has co-edited over 30 international COMADEM congress proceedings which have been published by major publishing organisations.
参会论文信息
Title：A Novel Residual Domain Adaptation Network for Intelligent Transfer Diagnosis
Author：Jinyang Jiao, Ming Zhao, Jing Lin, Kaixuan Liang, and Chuancang Ding
Abstract：Deep neural networks based intelligent diagnosis methods are able to learn powerful features for accurate fault classification, however they cannot always generalize well across changes in data distributions. To address this issue, a novel residual domain adaptation network is proposed for transfer diagnosis of machinery in this paper. In the proposed framework, one-dimensional residual network is designed as feature generator, then a mixed moment matching strategy, including first-order statistics and second-order statistics, is proposed to minimize the distribution discrepancy across domains. The comprehensive experiments on rolling bearing fault dataset are constructed to evaluate the proposed method. The results show the effectiveness of the proposed method.
参会论文信息
Title：Repetitive transient extraction algorithm for the fault diagnosis of planetary gearbox via encoder signal
Author：Chuancang Ding, Ming Zhao, Jing Lin, Kaixuan Liang, Jinyang Jiao
Abstract：This paper proposes a systematic framework for the fault detection and condition monitoring of planetary gearbox using internal encoder signal rather than traditional external vibration signal. In this work, the raw encoder signal is firstly converted into instantaneous angular speed signal through difference method. Then comb filtering is applied to remove the interferences and highlight the concerned feature components. Finally, sparsity-based signal decomposition algorithm is introduced to separate the fault transients and harmonic components. With the proposed method, the periodical fault transients are successfully extracted and the weak incipient fault can be effectively detected. Moreover, the validity of the proposed method is confirmed through the synthetic encoder signal and real data acquired from the planetary gearbox.
参会论文信息
Title：A tooth-wise dimensionality reduction approach based on encoder signal for the diagnosis of gearbox
Author：Kaixuan Liang, Ming Zhao, Chuancang Ding, Jinyang Jiao, and Jing Lin
Abstract：As the planetary gearboxes are widely used in the industrial applications, a novel method is presented to detect the anomaly of planetary gearbox. In this work, the data from a rotary encoder are analyzed to get the fault information. The proposed approach mainly divided into three steps. Considering that the adopted values are angular displacements, the tooth-wise samples are obtained from the measured signal firstly. Then all the samples are mapped into a high-dimension space by kernel PCA (PCA) to find the most discriminative dimensions. Finally, the similarity measurement is taken among the tooth-wise samples to reveal the anomaly. The validity of proposed method is demonstrated both simulated signal and experimental data.
