Search

3% Cr roller steel produced by two smelting process routes of the electroslag remelting and the rare earth refining respectively was studied， the chemical composition difference， microstructure， the characteristics of inclusion， as well as mechanical properties for these two roller steels were comparatively investigated by SEM， EDS， ASPEX etc characterization means and experimental methods of electrolysis extraction， tensile， impact tests etc . The results showed that w［Ce］ ≤0.01%， w［Al］ =0.022%， and w［Al］ =0.018% in the rare earth refined roller steel， and the two processes have similar control levels on the impurity elements in the steel. After heat treatment with the same process， there was similar microstructure of the tempered sorbite in these rare earth refining and electroslag remelting roller steel， and microstructure uniformity of roller body made by two processes was almost consistent. The inclusions in the rare earth refined roller steel are mainly ellipsoidal CeAlO3 and Ce2O2S the average size is about 4.17 μm， and the area percentage is about 0.0066%；Meanwhile some irregular Al2O3 particles about 3.88 μm in average diameter in the electroslag remelting roller steel were also observed about 0.008 1% in area fraction. Moreover， the yield strength of rare earth refining roller steel were few different from those of the electroslag remelting roller steel. But the elongation， area reduction and impact toughness of the rare earth refining roller steel were higher than those of the electroslag remelting roller steel.

Objective To establish a monitoring system and a standard database system for neonatal hearing screening program and evaluate its effect. Methods The hearing screening data of newborns in Nanjing in 2021 from the neonatal hearing screening information system of Jiangsu Province were selected as the experimental group, and the manual registration data of neonatal hearing screening in Nanjing in 2020 were selected as the control group. The initial screening rate, re-screening rate, referral rate and hearing impairment disease management rate of neonates were compared between two groups. Results The initial screening rate, re-screening rate and referral rate of neonatal hearing screening in the information management system were 99.27%,85.73% and 91.60% respectively, which were significantly higher than those in the manual registration control group (96.94%, 68.36% and 66.20%, P

The corrosion dissolution behaviour and influencing factors of 7000 series aluminum alloy dissoluble open - hole packers treated by electroless nickel plating and epoxy resin sealing in Tahe oilfield completion operation were analysed, and the fragments of dissoluble packers used for the first time was tested and analysed. Results showed that the well completion string configuration, wear during running in, medium in the well and high temperature and high pressure environment were the main factors prompting the rapid corrosion dissolution. Electroless nickel plating coating had certain protective effect on body, however,it brought the galvanic corrosion as well. Besides, the 7A04 aluminum alloy body and nickel plated aluminum alloy coupon were tested under simulated field conditions, and the corrosion rates in different media were obtained, which provided a basis for the prediction of the service life of aluminum alloy body and the optimization of coating thickness.

Deep learning based on neural network has achieved excellent results in a large number of fields,but it is difficult to deal with similar or untrained tasks,and it is difficult to learn and adapt to new tasks.Moreover,it requires a high scale of trai-ning samples,resulting in its poor generalization and expansion.Meta learning is a new learning framework,which aims to solve the problem that traditional learning methods can't solve fast learning and adapt to new tasks.Aiming at the meta learning problem of image classification,a novel fiber bundle meta learning algorithm based on Bayesian theory is proposed.Firstly,the convolution neural network is used to extract the image information supporting the dataset,and the image representation is obtained.Then the manifold structure of data features and the fiber bundle of data features are constructed.The input query set selects the manifold section of the current new task to obtain the fiber suitable for the new task,so as to get the correct label of the image.Experimental results show that the model based on the proposed algorithm (FBBML) achieves the best accuracy performance compared with the standard four-layer convolutional neural network model on the common data set (mini-ImageNet).At the same time,the fiber bundle theory is introduced into meta learning,which makes the algorithm more interpretable.