Search

Accurately measuring the current is crucial for assessing the operational status of the line. The tunnel transmission approach presents challenges due to its intricate line layout in limited space. Utilizing non-contact current measurement offers a superior alternative. To suppress the potential magnetic interference in the transmission tunnel, a novel M-shaped magnetic shielding design for current measurement is proposed. The structure is used to measure current in triangularly arranged conductors where magnetic interference exists at random position. First, a one-dimensional and two-dimensional sensor system based on a triangular arrangement of conductors are used to analyze the magnetic field. Then, the structure of the magnetic shielding is derived. The relationship between the geometrical parameters of shielding structure and the position of the sensors with mean shielding uncertainty (MSU) is discussed by finite element analysis (FEA) method. At current of 2000 A in the conductors, the lowest MSU for the M-shape shielding structure (0.19%) is significantly lower than the lowest MSU for the C-shape shielding (1.63%). Moreover, the M-shaped shielding fits better with the geometry of conductors. In addition, thickness of the shielding structure, position of shielding structure and position of sensors are optimized based on the Non-dominated Sorting Genetic Algorithm-II (NSGA-II). The FEA method is used to simulate magnetic field interference acting on a three-phase AC current with an amplitude of 2000A. The two-dimensional sensor system with shielding structure can reduce the root mean square error (RMSE) of current to be measured from 14.42% to 3.79%.

Deploying deep learning models to edge devices for low-latency and privacy-preserving applications has become a trend. To adapt to heterogeneous devices and data, it is significant to generate customized models. However, existing model adaptation approaches require edge devices to make interactions (collecting hardware information or local data) with the cloud, which raises privacy concerns, increases communication costs, and burdens the cloud. By contrast, we propose CamoNet, a universal on-device model adaptation framework with zero interaction between devices and the cloud. In CamoNet, a lightweight on-device neural architecture search module is utilized to quickly generate a customized model for subsequent on-device training, followed by an on-device contrastive transfer learning module to effectively leverage unlabeled data for fine-tuning the customized model. Extensive experimental results show that CamoNet can effectively run on various edge devices. Compared with the SOTA model adaptation approaches, CamoNet achieves significant accuracy improvement by 25.2% on average for image classification, 10.1% on average for object detection, and reduces the training memory by 4.8-11.4×. We will open-source our models and tools for edge AI developers.

Robot-assisted lobectomy (RAL) is increasingly used as an alternative to video-assisted lobectomy (VAL) for resectable non-small cell lung cancer (NSCLC). However, there is little evidence of any difference in postoperative health-related quality of life (HRQoL) between these two approaches.

It is necessary to study the influence mechanism of blade thickness on the photoacoustic signals since material thickness will affect defect detection precision by affecting the types, velocity, and frequency of photoacoustic signals. In this article, the influence mechanism of blade thickness on sound waves is obtained by setting defects with different depths in different thicknesses areas of blades. This mechanism instructs the imaging detection and characterization of defects with different depths in different thicknesses. First, by combining with the dispersion curves, we found that there are more anti-symmetric $A_{0}$ modes in thin area. Second, different types of sound waves lead to different velocities of sound waves. The sound velocity in the thin area of the blades is 2747.58 m/s, and the sound velocity in the thick area is 2912.22 m/s. Further, compared with the thick area, the thin area has more peaks and more complex high-frequency components, but the frequency range is basically the same. Finally, the detection and characterization of blade defects are carried out by combining imaging inspection and influence mechanism. The research results can not only promote the basic research on the influence of thickness on photoacoustic signals, but also expand the application of laser ultrasound in blade defect detection.

Cellular reprogramming can manipulate the identity of cells to generate the desired cell types1–3. The use of cell intrinsic components, including oocyte cytoplasm and transcription factors, can enforce somatic cell reprogramming to pluripotent stem cells4–7. By contrast, chemical stimulation by exposure to small molecules offers an alternative approach that can manipulate cell fate in a simple and highly controllable manner8–10. However, human somatic cells are refractory to chemical stimulation owing to their stable epigenome2,11,12and reduced plasticity13,14; it is therefore challenging to induce human pluripotent stem cells by chemical reprogramming. Here we demonstrate, by creating an intermediate plastic state, the chemical reprogramming of human somatic cells to human chemically induced pluripotent stem cells that exhibit key features of embryonic stem cells. The whole chemical reprogramming trajectory analysis delineated the induction of the intermediate plastic state at the early stage, during which chemical-induced dedifferentiation occurred, and this process was similar to the dedifferentiation process that occurs in axolotl limb regeneration. Moreover, we identified the JNK pathway as a major barrier to chemical reprogramming, the inhibition of which was indispensable for inducing cell plasticity and a regeneration-like program by suppressing pro-inflammatory pathways. Our chemical approach provides a platform for the generation and application of human pluripotent stem cells in biomedicine. This study lays foundations for developing regenerative therapeutic strategies that use well-defined chemicals to change cell fates in humans.

Supplemental Digital Content is available in the text

Human pluripotent stem-cell-derived islets (hPSC-islets) are a promising cell resource for diabetes treatment1,2. However, this therapeutic strategy has not been systematically assessed in large animal models physiologically similar to humans, such as non-human primates3. In this study, we generated islets from human chemically induced pluripotent stem cells (hCiPSC-islets) and show that a one-dose intraportal infusion of hCiPSC-islets into diabetic non-human primates effectively restored endogenous insulin secretion and improved glycemic control. Fasting and average pre-prandial blood glucose levels significantly decreased in all recipients, accompanied by meal or glucose-responsive C-peptide release and overall increase in body weight. Notably, in the four long-term follow-up macaques, average hemoglobin A1c dropped by over 2% compared with peak values, whereas the average exogenous insulin requirement reduced by 49% 15 weeks after transplantation. Collectively, our findings show the feasibility of hPSC-islets for diabetic treatment in a preclinical context, marking a substantial step forward in clinical translation of hPSC-islets.

Recently, we reported a chemical approach to generate pluripotent stem cells from mouse fibroblasts. However, whether chemically induced pluripotent stem cells (CiPSCs) can be derived from other cell types remains to be demonstrated. Here, using lineage tracing, we first verify the generation of CiPSCs from fibroblasts. Next, we demonstrate that neural stem cells (NSCs) from the ectoderm and small intestinal epithelial cells (IECs) from the endoderm can be chemically reprogrammed into pluripotent stem cells. CiPSCs derived from NSCs and IECs resemble mouse embryonic stem cells in proliferation rate, global gene expression profile, epigenetic status, self-renewal and differentiation capacity, and germline transmission competency. Interestingly, the pluripotency gene Sall4 is expressed at the initial stage in the chemical reprogramming process from different cell types, and the same core small molecules are required for the reprogramming, suggesting conservation in the molecular mechanism underlying chemical reprogramming from these diverse cell types. Our analysis also shows that the use of these small molecules should be fine-tuned to meet the requirement of reprogramming from different cell types. Together, these findings demonstrate that full chemical reprogramming approach can be applied in cells of different tissue origins and suggest that chemical reprogramming is a promising strategy with the potential to be extended to more initial types.Cell Research advance online publication 25 December 2015; doi:10.1038/cr.2015.142

Design and exploitation of efficient visible light photocatalytic systems for water splitting and degradation of organic dyes are of huge interest in the fields of energy conversion and environmental protection. Herein, two-dimensional CaIn2S4/g-C3N4heterojunction nanocomposites with intimate interfacial contact have been synthesized by a facile two-step method. Compared with pristine g-C3N4and CaIn2S4, the CaIn2S4/g-C3N4heterojunction nanocomposites exhibited significantly enhanced H2evolution and photocatalytic degradation of methyl orange (MO) activities under visible light irradiation. The optimal CaIn2S4/g-C3N4nanocomposite shows a H2evolution rate of 102 μmol g–1h–1, which is more than 3 times that of pristine CaIn2S4. The mechanisms for improving the photocatalytic performance of the CaIn2S4/g-C3N4nanocomposites were proposed by using the photoluminescence measurement and electrochemical analyses. It was demonstrated that the enhanced photocatalytic performance of CaIn2S4/g-C3N4heterojunction nanocomposites mainly stems from the enhanced charge separation efficiency. In addition, a plausible mechanism for the degradation of MO dye over CaIn2S4/g-C3N4nanocomposites is also elucidated using active species scavenger’s studies.

In this paper, we discuss the emission of visible light by a monolithically integrated silicon gate-controlled diode with the p-n junction reverse-biased. Since the MOS-like diode utilizes the field effect to modulate the optical output, the modulation speed will benefit from this mechanism. Hence, a silicon gate-controlled diode structure for optical modulation analyzed its modulation principle, its dynamic characteristics are presented, and the bandwidth of the device is considered to approach GHz in theory due to the field-induced emission mechanism. The prototype MOS-like diode opens up the design of multiterminal silicon light emitting devices (LEDs), where gate electrodes with more than one gate contact overlap several junctions with different junction intersection geometries. The device appears as a good candidate for optical modulation within silicon technology.

Nowadays, emission reduction becomes the main direction of energy system development. Replacing fossil fuel with renewable energy in energy sectors such as the power and transport sector serves as one of the essential measures. However, scientific gaps appear in how this renewable transition can be reached in an energy system, including the power and transport sector under the goal of “carbon neutral” in China towards 2060. Taking Sichuan province in China as an example, this study establishes the renewable transition pathways in an energy system including the power and transport sector and evaluates the feasibility of different renewable technologies in the two sectors. Three renewable scenarios based on various renewable electricity in the power sector and three renewable scenarios based on the electrification degree in the transport sector are formulated in the EnergyPLAN tool. The results show that all the renewable scenarios can reduce the emissions to zero in Sichuan in 2060, with the CO2emissions in the business as usual (BAU) scenario reaches 135.7 Mt. Moreover, adding wind power to its maximum potential has better capacity to meet the requirement of further electrification in the transport sector than solar power in Sichuan province. From the economic perspective, all the renewable scenarios have similar energy system cost with the BAU scenario, reaching a range of 203 and 244 BCNY/y. All in all, the methodology and results in this study contribute to the policy-makers for constructing a carbon-neutral energy system.

Vibration stimulation was used to produce light/conventional oil but has seldomly been used in heavy oil and bitumen production. As an innovative enhanced heavy oil recovery technology, the vibration-stimulated gas pressure cycling (VS-GPC) process, for the first time, is proposed in this research. A series of gas pressure cycling (GPC) and VS-GPC tests are performed on a reliable vibratory testing system specially designed and assembled for the study. The experimental data quantifies the impact of the externally applied vibration on improving the heavy oil recovery with cyclic gas/solvent injection. The effects of the vibration amplitude, frequency, and duration on the heavy oil recovery are investigated systematically. The detailed phenomenon of vibration stimulation on fluid mobility and flow is explored in porous media. It is found that the lowest frequency of vibration carried results in the highest heavy oil recovery among the CO2-VS-GPC tests completed. This research provides a guide for designing a practically sound technique for enhanced heavy oil recovery with the combination of GPC and externally applied vibratory stimulations.