Search

Harmonic blocking is usually applied by converter transformer differential protection to prevent protection mal-operation during an inrush condition, but it may incorrectly block the protection during an internal fault. This paper first makes an in-depth analysis on factors that lead to differential protection failure to trip for converter transformer internal faults. The main factor responsible for the protection failure to trip is half-cycle saturation of the converter transformer core, which results from the dc component in the fault current. This suggests that harmonic blocking may not be suitable for differential protection of converter transformers to identify inrush conditions. In this regard, this paper proposes a novel method using mathematical morphology to effectively discriminate between the conditions caused by inrush and internal faults. This method implements the waveform symmetry criterion and improved morphological gradient criterion to distinguish between these two conditions. The effectiveness of the proposed method is evaluated based on extensive simulation, and the presented results show that the proposed method can ensure operation for all types of internal faults. Additionally, this method improves the accuracy of inrush identification, even in extreme cases, such as sympathetic inrush and inrush with current transformer saturation.

Background and aims Atherosclerosis, a progressive inflammatory disease characterized by elevated inflammation and lipid accumulation in the aortic endothelium, arises in part from the infiltration of inflammatory cells into the vascular wall. However, it is not fully defined how inflammatory cells, especially macrophages, affect the pathogenesis of atherosclerosis. Schlafen4 (Slfn4) mRNA is remarkably upregulated upon ox-LDL stimulation in macrophages. Nonetheless, the role of Slfn4 in foam cell formation remains unclear. Methods To determine whether and how Slfn4 regulates lesion macrophage function during atherosclerosis，we engineered ApoE–/–Slfn4–/– double-deficient mice on an ApoE–/– background and evaluated the deficiency of Slfn4 expression in atherosclerotic lesion formation in vivo. Results Our results demonstrate that total absence of SLFN4 and the bone marrow-restricted deletion of Slfn4 in ApoE–/– mice remarkably diminish inflammatory cell numbers within arterial plaques as well as limit development of atherosclerosis in moderate hypercholesterolemia condition. This is linked to a marked reduction in the expression of proinflammatory cytokines, the generation of the reactive oxygen species (ROS) and the apoptosis of cells. Furthermore, the activation of MAPKs and apoptosis signaling pathways is compromised in the absence of Slfn4. Conclusions These findings demonstrate a novel role of Slfn4 in modulating vascular inflammation and atherosclerosis, highlighting a new target for the related diseases.

The patent laws of many countries have ''research exemption'' provisions that exempt certain research-related uses of proprietary materials from patent infringement. By limiting the rights of existing patent holders, such rules are meant to facilitate follow-on innovation and benefit latecomer firms, especially in the pharmaceutical industry. In this paper, we provide the first study of the impact of the research exemption, exploiting unique features of the institutional setting in China. Using firm-level data from 2007 to 2018 and a difference-in-differences (DID) strategy, we find that the research exemption in China leads to a large increase in firms' R&D inputs. However, there is no evidence that the research exemption leads to more patents or improved productivity on average. Further analysis reveals that it only causes an increase in patents for few firms with large market power. Overall, our findings highlight the importance of understanding the relationship between firms' innovation and imitation strategies for the design of innovation policies.

In this paper, we empirically study the impact of gender identity on the wife's labor supply in China where a high female employment rate and a large population that has a traditional gender identity coexist. We find that the wife's gender identity affects her labor force participation for all cohorts, and the wife's gender identity affects earnings only for rural cohorts. The impact of gender identity varies among generational, regional, and educational cohorts. Husbands still guide their wives, i.e., a husband's gender identity affects his wife's labor force participation and earnings.

Facial landmark detection in uncontrolled natural condition is a challenging task due to illumination, occlusion, expression, posture, blur and other factors. In order to make a stable and accurate prediction, a novel data augmentation strategy based on random mask is used to obtain an efficient training data set. Our work is motivated by the ability of artist’s spatio imagination from part to whole. The artist can find the corresponding key points according to the incomplete image, so as to realize the whole from the part. In the experiment, the corresponding incomplete image is first generated based on the random mask and the complete image, and then the incomplete image is fed into the neural network to fit the mapping between the incomplete image and the facial landmarks. In order to evaluate the performance, a simple regression convolutional neural network based on the improved VGG16 [1] was trained, and the latest technical results were obtained. Furthermore, our method is tested over both complete and incomplete face images. Extensive experiments reveal that our model can work well on different human face data set.

Converter transformers in high-voltage direct current (HVDC) systems are often subjected to DC bias induced by earth return currents (ERCs) from HVDC systems or by geomagnetically induced currents (GICs). DC bias may adversely affect the performance of converter transformers, but its impact on the protection of converter transformers remains uncertain. This paper thoroughly investigates the response of converter transformer differential protection under the effect of DC bias. The network was simulated in PSCAD/EMTDC, and the modeled components included the converter transformers, the current transformers (CTs), and the DC bias signatures. The fault currents generated by the simulator were downloaded into a numerical relay and its operating behavior was observed. The results indicated the pre-existing harmonics, induced by DC bias, do not cause harmonic blocking to incorrectly inhibit the relay for in-zone faults; and any spurious trip, caused by DC bias, can be prevented by appropriate harmonic blocking and the use of a higher slope 2 in the differential current versus bias current characteristic. It was demonstrated the low-frequency characteristic of GIC did not pose any new threats to differential protection. This indicates the low-frequency characteristic of GIC is negligible when investigating the impact of GIC.

To improve the output power and electron efficiency and broaden the operating bandwidth, the three-gap extended output cavity was used in the relativistic klystron to replace the single-gap output cavity. The high-frequency characteristics of the three-gap extended cavity were studied. The electron load conductance was derived and corrected based on the theory of relativity, by which a more accurate relation of electron load conductance versus transmit angle can be obtained. The characteristic impedance is three times bigger than that of the single-gap cavity under the case of the 2 pi mode, which can improve the M-2(R/Q) and the beam-wave interaction efficiency. The model of sheet beam klystron (SBK) is comprised of three-gap input cavity, two-gap first idler, three-gap second idler cavity, and three-gap output cavity. The 3-D particle-in-cell simulation results show that for a 500 kV, 1 kA, 40.134 GHz, and 20 W input power, the SBK is capable of generating an output power higher than 142 MW with the gain of 68.5 dB, maximum interaction efficiency of 28.4%, and 3-dB bandwidth of 240 MHz.

Packaging engineering, one of engineering programs, which is characterized by an interdisciplinary specialty, has been founded in China since the early 1980s. It has well been known that strengthening the construction of its teaching team is of great significance of pushing on the intensive and characteristic development, promotion of teaching quality, as well as the aggressive adaptation of the new normal of higher education. In this paper, taking a comprehensive reform of packaging engineering program of our University as an instance, some systems and practices on the construction of teaching team are focally introduced, inclusive of moral guidance, system and incentives, overall deepening of teaching innovation.

On the basis of comprehensively surveying the background and current situation of bilingual teaching implementation at institutions of higher learning, China, the distinct difference between bilingual courses and academic English for packaging in terms of bilingually educational model is clearly defined. It is emphasized that bilingual teaching in packaging engineering should protrude such kernel elements as packaging knowledge, and heighten the level of students who are able to apply foreign languages to carry on communications in discipline and specialty. In the meanwhile, the implementation of bilingual teaching ought to first and foremost be sustained, and how to foster the packaging professionals with international horizon and competitiveness is dwelled on. Ultimately, it is recommended that the innovative bilingual teaching model should be explored to accommodate the colleges and specialties concerned, inclusive of these key factors like teaching team, curriculum, teaching contents and resources, teaching means and methods, and the acceleration of the pace of course construction is taken as an important starting point and the connotative development of bilingual teaching is really implemented to elevate the quality of bilingual teaching in packaging engineering through deepening innovation of bilingual teaching.

Video super-resolution (VSR) has become more and more important in vision perception. Currently, most of video super-resolution methods used optical flow estimation by explicit motion compensation. We proposed a new network using multi-frames as input without motion compensation. In order to improve the performance of the reconstructed high-resolution (HR) frame, perceptual loss function is used to evaluate the deviation between the prediction and ground truth. Perceptual loss characteristic is able to improve the quality of image reconstruction on the vision, but with a low Peak Signal to Noise Ratio (PSNR). Experimental results show that compared with other methods, our model can recover more details in high-resolution image, although it does not get higher PSNR value.

Most of learning-based super-resolution (SR) methods reconstructed high resolution (HR) image using a very deep convolutional neural network (CNN). CNN-based methods are used to process low resolution images, the predicted image is too smooth and some texture details are ignored. In order to improve the quality of the recovered high resolution image, wavelet transform is applied at the end of our model. Firstly, the prediction image is obtained through the CNN model; then, the predicted image and the ground truth image are processed into four frequency channels by wavelet transform; at last, four frequency channels and original image are considered when calculating loss. Compared with the original CNN model, the results of adding wavelet transform is obviously improved.

I. INTRODUCTION The existence of liquidity constraints remains a significant issue within intertemporal models of household choice. As Hayashi [1985b], Cox and Jappelli [1993], and other studies detail, the notion that consumers may be unable to borrow all they desire to finance current consumption carries important implications pertaining to Ricardian equivalence as well as the dynamic effects of government spending versus tax policies. Liquidity constraints have also been advanced as a primary reason for empirical rejections of the Permanent Income/Life Cycle Hypothesis, as discussed in Zeldes [1989]. Testing for the existence of debt constraints with micro data has generated mixed results.(1) Hall and Mishkin [1982], Mariger [1987], Zeldes [1989], Hayashi [1985a], and Duca and Rosenthal [1994] argue for the significance of liquidity constraints. Jappelli [1990] and Duca and Rosenthal [1994] contend that a nontrivial segment of the population faces credit rationing, based upon information from the Survey of Current Finances. On the other hand, the findings of Altonji and Siow [1987], Runkle [1991], Mariger and Shaw [1993], and Attanasio and Weber [1995] provide little evidence of liquidity constraints. This article investigates liquidity constraints with endogenous income. It centers around the household's intertemporal model with nonseparable consumption and leisure. Following the methodology of Zeldes [1989] and Runkle [1991], the formulation allows for an individual household at a point in time to either face binding or nonbinding liquidity constraints. Our model extends this work by considering labor supply within the family's choice set, as recommended by Hayashi [1985a] and Attanasio and Weber [1995]. It includes endogenous wage income in the minimum wealth constraint. These features allow for consideration of work-related decisions within household utility maximization under liquidity constraints. For example, a family may choose increased labor supply to remove themselves from a debt constraint or to enable them to apply for a larger loan. This behavior may take the form of increased overtime, finding a second job, or having the spouse enter the labor market. The model works toward remedying several difficulties associated with empirical testing for liquidity constraints. Following Alessie, Melenberg, and Weber [1988], we derive from the Euler equations an estimable equation for employed households. The equation holds for both liquidity constrained families as well as households that do not face a binding debt constraint. This result addresses a well-known problem involved with the standard consumption-based utility formulation, as discussed by Hayashi [1985b] and others, the inability to directly estimate the Euler equation(s) for liquidity constrained households. Consequently, work such as Hayashi [1985a], Zeldes [1989], and Runkle [1991] use an a priori criterion to split the sample into suspected liquidity constrained and non-liquidity constrained groups. Their tests are based upon parameter estimates from the group not considered to be debt constrained. Given estimated structural parameters from the consumption-leisure model, we can obtain estimates related to the individual Lagrange multipliers pertaining to the minimum wealth constraint. These estimated data series enable us to test directly for the presence of credit constraints within the general population. Primarily due to the difficulty in estimating the Lagrange multipliers, indirect means have generally been used to test for debt constraints over a sample of households. Hall and Mishkin [1982] base their inferences on the excess sensitivity of consumption forecast errors to income innovations. Mariger and Shaw [1993] raise doubts regarding their findings, showing that such tests of the Permanent Income-Life Cycle hypothesis can be biased toward rejection. Zeldes [1989] and Fissel and Jappelli [1990] criticize the general method, since it assumes a constant proportion of liquidity constrained households. …

It is an important target that high quality application-oriented talents with self-employed consciousness and innovation ability are trained via promoting quality-oriented education. Hence, the study serves curricula construction as the platform, sets up scientific and reasonable curriculum system, strengthens practical teaching contents and reforms teaching methods on the basis of professional features. Consequently, the transform from attention of knowledge-transference to emphasis on ability training is achieved, in the meantime, innovative and entrepreneurial capacities of packaging talents are improved.

The different senses between Bilingual Teaching and English for Special Purpose are emphasized. The construction of Packaging Technology, a National Bilingual Teaching Demonstration Course, is specified. The educational model of Packaging Engineering Students is proposed, which will meet the international requirement on packaging education. Regional universities and colleges can refer to this study for the internationalization of cultivating professionals.

Peer-Instruction (PI) pedagogy focuses on self-access study, cooperation, communication and interaction, the critical factors of the PI pedagogy are how to stimulate the consciousness of students and guide students to explorative study. In our bilingual course, ie., Packaging Technology, which was awarded one of the national bilingual teaching demonstration courses in 2010, the PI concept is introduced, in which teachers play key role in controlling and guiding in teaching. Such teaching methods as heuristics, participation and research-based learning are adopted to stimulate students’ ability in self-access and improve the quality of bilingual teaching.

An X-band coaxial multi-beam relativistic klystron amplifier is designed in order to increase output microwave power and operating frequency of the amplifier tube. The experiment is performed on a Tesla-type accelerator. The amplifier is driven by an electron beam of 2.8 kA at 720 kV, and a microwave power of 30 kW and frequency of 9.384 GHz is injected into an input cavity by means of an external source, then a microwave power of over 800 MW is extracted, the amplifier gain is about 44 dB, and conversion efficiency is 40%. The experiment proves that output power of nearly GWs can be generated with the X-band coaxial multi-beam relativistic klystron amplifier driven by a kW-level input power.