Search

The financial system has reached its pinnacle because of economic and social growth, which has propelled the financial sector into another era. Public and corporate financial investment operations have significantly risen in this climate, and they now play a significant part in and impact the efficient use of market money. This finance sector will be affected by high-risk occurrences because of the cohabitation of dangers and passions, which will cause order to become unstable and definite financial losses. An organization’s operational risk is a significant barrier to its growth. A bit of negligence could cause the business’s standing to erode rapidly. Increasing funding management and forecasting risks is essential for the successful development of companies, enhancing their competitiveness in the marketplace and minimizing negative effects. As a result, this study takes the idea of mobile edge computing. It creates an intelligent system that can forecast different risks throughout the financial investment process based on the operational knowledge of important investment platforms. The CNN-LSTM approach, based on knowledge graphs, is then used to forecast financial risks. The results are then thoroughly examined through tests, demonstrating that the methodology can accurately estimate the risk associated with financial investments. Finally, a plan for improving the system for predicting financial risk is put out. [ABSTRACT FROM AUTHOR]

Proteins are the building blocks of life, carrying out fundamental functions in biology. In computational biology, an effective protein representation facilitates many important biological quantifications. Most existing protein representation methods are derived from self‐supervised language models designed for text analysis. Proteins, however, are more than linear sequences of amino acids. Here, a multimodal deep learning framework for incorporating ≈1 million protein sequence, structure, and functional annotation (MASSA) is proposed. A multitask learning process with five specific pretraining objectives is presented to extract a fine‐grained protein‐domain feature. Through pretraining, multimodal protein representation achieves state‐of‐the‐art performance in specific downstream tasks such as protein properties (stability and fluorescence), protein‒protein interactions (shs27k/shs148k/string/skempi), and protein‒ligand interactions (kinase, DUD‐E), while achieving competitive results in secondary structure and remote homology tasks. Moreover, a novel optimal‐transport‐based metric with rich geometry awareness is introduced to quantify the dynamic transferability from the pretrained representation to the related downstream tasks, which provides a panoramic view of the step‐by‐step learning process. The pairwise distances between these downstream tasks are also calculated, and a strong correlation between the inter‐task feature space distributions and adaptability is observed. [ABSTRACT FROM AUTHOR]

Parasitic mutual inductance between capacitors is critical for the high-frequency performance of electromagnetic interference filter. In this article, a new cancelation method is proposed to eliminate mutual inductance completely without adding additional components. The principle of the cancelation method is to design the layout of input and output traces, and the mutual coupling of traces can be equivalent to adding a negative inductance on the transfer branch of the filter to cancel the mutual inductance between capacitors. The two cancelation designs of the filters with small and large capacitor spacing are proposed, respectively. Then, the filter model is established in the finite-element method analysis tool, and the mutual inductances under different layout parameters are calculated to explore the recommended layout, which is to make the negative mutual inductance of traces fully offsets the positive mutual inductance between capacitors. Finally, the effectiveness of cancelation method and the accuracy of model are verified by experiments. The method can significantly improve the filter performance without extra costs and is expected to become a filter design that can be widely promoted. [ABSTRACT FROM AUTHOR]

With the rise of deep learning technology, due to the superior performance of the deep neural network, its application in the digital economy has attracted extensive attention of scholars. Since the beginning of the 21st century, my country's digital economy has developed rapidly, and its universalization and other characteristics have created favorable conditions for the optimal allocation of resources in underdeveloped regions and the exertion of comparative advantages. The digital economy will play a key role in poverty alleviation, promoting coordinated regional development, narrowing regional gaps, and improving the spatial layout of my country's reform and opening up. This paper studies the factors that the digital economy based on deep neural networks has on regional economic growth. Simulation experiment conclusions are as follows: (1) the digital economy of Guizhou, Beijing, Chongqing, Anhui, and Tibet is growing rapidly. The central and western regions are in a period of rapid growth. For the gap between major industrial provinces, the coefficient of variation reached about 1 before 2013, then it declined rapidly, and slowed down, and steadily declined after 2019, indicating that the gap in the digital economy in various regions is narrowing in general. (2) From the national level, the digital economy index coefficient is 1.24, that is, for every 1% increase in digital economy investment, GDP will increase by about 0.24%. The labor force increased by 1% and the GDP increased by about 0.22%. This promotion effect is also very obvious. (3) Judging from the above data, the western region urgently needs to promote the construction of the digital economy and introduce high-tech digital economy talents. The talent effect in the Midwest has a significant effect on GDP. (4) From the perspective of the whole country and other regions, the parameter coefficients and signs have not changed significantly, so the original model is robust, and so the conclusion is desirable. [ABSTRACT FROM AUTHOR]

Insulated gate bipolar transistor (IGBT) has been widely used in the voltage-source-converter-based high-voltage direct current (VSC-HVDC). However, the electromagnetic interference (EMI) generated by its fast switching process affects the normal operation of the surrounding power equipment. The switching voltage and current of IGBT are the main sources of EMI. Different driving circuit parameters result in different IGBT switching characteristics. In this article, the switching model of IGBT is introduced first and the influence of driving parameters on EMI sources is analyzed. Based on the conducted EMI experiment platform of an IGBT module, the time-domain waveforms of switching voltage and current under different driving parameters are obtained and compared with the model. Then the interference voltage spectrums are measured by an EMI receiver under different driving parameters. The influence of driving parameters on conducted EMI in 0.15–30 MHz is then analyzed and explained by combining the result of time domain and frequency domain. This article comprehensively analyzes the relationship between driving parameters and EMI sources and summarizes the influence characteristics of driving parameters on common-mode (CM) and differential-mode (DM) noise. The conclusions can help the parameters design of driving board, which can effectively reduce the conducted EMI of the converter. [ABSTRACT FROM AUTHOR]

This article addresses the influence of parameters spread on transient current distribution among parallel-connected silicon carbide (SiC) mosfets and proposes a chips classification method to suppress current imbalance. A comprehensive comparison of parameter spread between silicon (Si) and SiC mosfets is first presented. Then, a new classification criterion, referred to in this article as a distance coefficient of transfer curves (DCTC), is proposed to characterize device spread. In addition, the sensitivity analysis of transient current imbalance (TCI) to device spread is carried out. It is found that TCI increases almost linearly with increasing DCTC. Furthermore, a hierarchical cluster algorithm is developed to achieve an automated chips classification for multiple devices intended for paralleled application. This algorithm may as well facilitate the chip selection process for multiple-chip power module packaging. Moreover, the influence of operating temperature on classification is also discussed. It should be noted that chips classification needs to consider real application conditions and package parasitic. Finally, a test bench with a round layout is designed, to keep circuit layout asymmetry to a minimum, and used to experimentally verify the performance of the classification method. The experimental results validate the effectiveness of the proposed classification method for suppressing TCI. [ABSTRACT FROM AUTHOR]