Electronics Science Technology and Application (E-Journal)
Not a member yet
337 research outputs found
Sort by
A New Structure Bipolar Plate for Vanadium Flow Battery
Get PDFA pattern bipolar plate (BP) is designed to decrease the electrical contact resistance between the BP and carbon felt electrode (CFE) by increasing the BP's specific surface area through hot pressing the spherical crown pattern on the surface of the BP, and the carbon felt is pressed into the pattern by the stack tightening force. The contact area growth rate is 18.3% of a single cell which has 48cm2 active area. The areal specific resistance (ASR) of conventional BP and pattern BP is 1309 mΩ·cm2 and 946 mΩ·cm2 at 20% electrode compression ratio, the pattern plate is 27% lower than the conventional plate. In the single battery charge/discharge cycle test, battery assembled with the pattern BP shows a 1% higher EE than the conventional BP
Application of Deep Learning in Computer Vision
Get PDFEvery aspect in society in the process of development towards intelligence, automation, the role of artificial intelligence and the contribution is not allow to ignore, and computer vision technology is the key technology of artificial intelligence, with continuous innovation and development in the field of computer vision, the deep learning approach will be more and more widely applied in the field of computer vision, This paper analyzes and expounds deep learning, computer vision and deep learning framework respectively, and then focuses on image analysis, target detection and attitude estimation to effectively explore the application of deep learning in the field of computer vision, and carries out deep learning classification network research for reference
From the Perspective of Robotic Research and Development in Medicine: Analysis of Principles for the Application of Circulating Neural Networks and the Design of Economic Products in Biomedical Medicine
Get PDFThe circulatory neural network is an important component part of the artificial neural network and can be integrated into biomedical engineering for disease warning in current frontier research. In the current market economy, women's sexual pleasures can be designed using the circulatory neural network as a core biomedical engineering device. We performed technical analysis, expound the principles of neuroscience, and report and analyze them
Particle Swarm Algorithm to Optimize LSTM Short-Term Load Forecasting
Get PDFAccurate load forecasting is of great significance for national and grid planning and management. In order to improve the accuracy of short-term load forecasting, an LSTM prediction model based on particle swarm optimization (PSO)algorithm is proposed. LSTM has the characteristics of avoiding gradient disappearance and gradient explosion, but there is a problem that parameters are difficult to select. Therefore, particle swarm optimization algorithm is used to help it select parameters. The experimental results show that the optimized LSTM has higher prediction accuracy
CNN-Based End-to-End Unmanned Driving Simulation in Curved Scene
No full textNowadays, the research on autonomous driving technology is more and more in-depth, and the purpose of the research is that autonomous vehicles can achieve safe driving in various driving scenarios. In order to improve the existing autonomous driving scene data set, this paper proposes an end-to-end scene reality simulation implementation method based on CNN. This method is used to train an end-to-end model based on vision sensors for cornering scenes, establishing a direct mapping relationship between vision sensors and driving behavior. In this experiment, the Udacity simulator autonomously collects the data set of the curve scene, and after cutting and normalizing the original data, an end-to-end model with a low loss value is trained, and then the image data is divided into boxes. , Gaussian, median and bilateral filtering operations, the results show that filtering the data can quickly reduce the network loss value. Finally, comprehensive analysis can prove that this method is a high-performance and efficient scene implementation method
Research Progress of All-In-One Supercapacitor
Get PDFWith the development of flexible electronics, supercapacitors have received great attention due to their own advantages, of which the rapidly developing all-in-one supercapacitors have received more favour from researchers in recent years. Because they not only can reduce the interfacial resistance between electrolyte and electrode, but also ensure stability in external deformation. This paper provides a comprehensive review of all-in-one supercapacitors. Firstly, the similarities and differences between conventional and all in one supercapacitors are compared, followed by a discussion of the three main assembly methods for all-in-one supercapacitors
Development and Implementation of Software for Multi-Algorithm Image Quality Enhancement
Get PDFDifferent image defogging methods should be adopted for different degrees of low contrast and sharpness of images caused by different weather conditions. This paper builds a MATLAB-based image quality improvement and evaluation software that combines the global RGB histogram equalization algorithm, the global HSV histogram equalization algorithm, the restricted contrast adaptive histogram equalization algorithm, the single-scale Retinex algorithm, the multi-scale Retinex algorithm, the Multi-Scale Retinex with Color Restoration, and the dark channel a priori algorithms and their optimization algorithms for image quality improvement and evaluation software. Outdoor images of hazy days, rainy days and snowy days are selected and the best algorithms for different weather conditions are obtained through extensive experimental simulations, software processing and analysis
Research on Big Data Technology System of Smart City
Get PDFWith the coming of the age of intelligence, the new information and communication technology represented by big data technology brings new perspectives and foundations to solve the traditional city problems. Based on the actual needs of urban smart development and urban governance, big data should be excavated and processed reasonably and effectively, and the supporting role of big data for urban development should be studied, and a more practical strategic development path has been proposed. Based on the research of the current smart big data technology system, this article analyzes facts, pinpoints difficulties, and proposes solutions to problems, and contributes to the modernization of the national governance system and governance capabilities
Computer and Mathematical Modeling: Translational Research and Economics in Clinical Diagnostics
Get PDFThe computer and modeling approach has begun to be used extensively in clinical intelligence diagnosis, we have refined the necessary techniques related to intelligence medicine, and we have performed economics-directional analysis of models and structures of artificial intelligence in the translational medicine sense.At the same time, the development of clinical diagnostic techniques is also the result of constant innovation, and we propose the necessary strategy for a cross-disciplinary approach to clinical diagnostics and computer and mathematical modeling, with the authors reporting in conjunction with the results of the study
Anomaly Detection and Prediction Based on Holt-Winters Method
Get PDFOutliers, detection and prediction of abnormal periods, and prediction of trends in a given time base on existing data are the first problems to be solved in intelligent operation and maintenance. This paper took KPI performance index of 58 cells covered by 5 base stations from August 28 to September 25, 2021 as research data, chose 3 indexes (average number of users, PDCP traffic and average activation number), established a set of outlier prediction system based on Holt-Winters method