Machine fault diagnosis with small sample based on variational information constrained generative adversarial network

In actual engineering scenarios, limited fault data leads to insufficient model training and over-fitting, which negatively affects the diagnostic performance of intelligent diagnostic models. To solve the problem, this paper proposes a variational information constrained generative adversarial network (VICGAN) for effective machine fault diagnosis. Firstly, by incorporating the encoder into the discriminator to map the deep features, an improved generative adversarial network with stronger data synthesis capability is established. Secondly, to promote the stable training of the model and guarantee better convergence, a variational information constraint technique is utilized, which constrains the input signals and deep features of the discriminator using the information bottleneck method. In addition, a representation matching module is added to impose restrictions on the generator, avoiding the mode collapse problem and boosting the sample diversity. Two rolling bearing datasets are utilized to verify the effectiveness and stability of the presented network, which demonstrates that the presented network has an admirable ability in processing fault diagnosis with few samples, and performs better than state-of-the-art approaches.     

Modelling dislocation assisted tempering during rolling contact fatigue in bearing steels

Rolling contact fatigue in bearing steels is manifested by dark-etching regions, which are attributed to deformation induced tempering. In order to quantitatively explain this phenomenon, a model is suggested for martensite tempering assisted by dislocation glide during rolling contact fatigue. In the model, dislocations transport carbon from the matrix to carbide particles, provided that the carbon is located at a certain distance range from the dislocation contributing to the tempering process. By calculating the amount of carbon in the matrix, the kinetics of carbide thickening and hardness reduction are computed. It is found that the dark-etching region kinetics can be controlled by both bearing operation conditions (temperature and deformation rate) and microstructure (type, size, and volume fraction of carbides). The model is validated against tested bearings, and its limitations are discussed.     

On the thermally-induced seizure in bearings: A review

This paper presents a state-of-the-art survey of papers reported on the nature of a troublesome failure mode in bearings known as seizure. This mode of failure is thermally-induced and it occurs in both journal and rolling element bearings. To gain insight, particular attention is given to reported experimental observation, various mechanisms involved, and available prediction methodologies.     

攀钢1450热轧带钢厂精轧机组电动压下机构带钢压下时的动特性计算和探讨

从理论上和实践上对武钢１７００、鞍钢１７００改造前后、本钢１７００改造前后和攀钢１４５０精轧机组的电动压下机构在带钢压下时的动特性进行计算和探讨，得出有说服力的结论。主题词：     

中厚板轧制边部不均匀变形机理的探讨

通过分析宽中厚板轧制过程中的边部不均匀变形，探讨了其影响因素及大宽度板“切边纵裂”的产生原因，提出了解决问题的具体办法。     

ZJ510L-B汽车用高强度热轧板的试制

珠钢150t电弧炉-LF(VD)-CSP(紧凑式带材生产)流程试生产的5～8mm ZJ510L-B(0．16C--1．22Mn)汽车用高强度热轧板的铁素体晶粒尺寸为6．0～9．2μm，带状组织为1．5～2．5级，强度极限σb562～610MPa，延伸率δ5 32．0％～38．6％。文中还分析和提出了带状组织的形成原因和改进措施。     

厚度控制技术在中厚板轧制生产中的应用

本文探讨了影响中厚板厚度波动的因素，结合安钢中板厂2800mm轧机按合同轧制一批交货重量不大于理论重量船板的生产实践，总结了实际生产中实现高精度厚度控制的有效方法。     

微合金化高强度钢轧制采用的工艺技术

本文综述微合金化高强度钢轧制采用的工艺技术，其中包括热装热送、板坯加热、控轧控冷、快速冷却、采用的新设备等工艺技术。重点是生产低碳、高韧性、高强度、易焊接的造船用钢采用形变热处理技术，以提高产量和质量，节省昂贵的Ｎｉ、Ｃｒ、Ｍｏ等合金，降低生产成本。     

新型线圈式磁弹性扭矩传感器

阐述了新型线圈式磁弹性扭矩传感器的原理,推导出其输出表达式,进而说明了线圈式磁弹性扭矩传感器具有输出平稳、响应快的特点。     

��������ʵ�ֶ�������ϵ��Դ�����DZ���Ϳ�����

中国重庆川东地区是四川盆地天然气的主要储产地区，对川气东输工程具有举足轻重的影响，但由于勘探难度增大，使储量增长速度明显放慢，储采比快速下降，若元新的发现，将无法保证长期高产稳产。章针对该地区天然气储量资源的结构及现状，分析了二叠系、三叠系动勘探开发的潜力，认为在该区实施二叠系、三叠系滚动勘探开发是一项投资省、见效快的工作，指出重庆气矿近斯的发展方向应针对二叠系、三叠系实施滚动勘探开发。章还提出了相应的技术保障措施和近期工作的初步设想，优选了近期工作目标。