In the era of modern technology, machine learning and natural language processing has been adopted to be applied in several application areas. Natural language processing consists of diversified techniques such as tex...
详细信息
the proceedings contain 31 papers. the topics discussed include: large-signal modeling for full-bridge LLC resonant converter using extended hyperbolic tangent function;portfolio strategy of power producer considering...
ISBN:
(纸本)9781665493277
the proceedings contain 31 papers. the topics discussed include: large-signal modeling for full-bridge LLC resonant converter using extended hyperbolic tangent function;portfolio strategy of power producer considering energy storage in spot market, ancillary market and option market;multi-period two-stage stochastic der aggregation for power flexibility reserve;distributed photovoltaic-storage system optimization planning considering flexible resources in smart distribution network;echo state network based noise detection in energy internet orienting justice blockchain data;joint bidding strategy of onsite energy storage subject in multi-market collaborated with wind power plants;intelligent control technology of air source heat pump for mango drying;a bidirectional converter stabilization control strategy for novel power systems;and analysis of carbon emission reduction potential of regional multi-energy system considering the influencing factors of carbon emissions.
the proceedings contain 87 papers. the topics discussed include: misinformation detection in online social networks using content information;cybersecurity and IoT: open challenges and perspectives;an information cent...
ISBN:
(纸本)9781728129464
the proceedings contain 87 papers. the topics discussed include: misinformation detection in online social networks using content information;cybersecurity and IoT: open challenges and perspectives;an information centric networking approach to publish-subscribe in mobile IoT systems;social network analysis. why is it needed? how to do it? what can we expect from it?;combining edge and cloud for smart cities applications;comparison of lexicon performances on unstructured behavioral data;identifying influencers using time series analysis;an improvement proposal of genetic algorithms based on information entropy and game theory;and identifying concerned citizen communication style during the state parliamentary elections in Bavaria.
CubeSat systems are preferred for short-term space missions, like those with a scientific objective, since they make it possible to complete them in an affordable cost. On the other hand, the resource-constrained natu...
详细信息
ISBN:
(纸本)9781450397407
CubeSat systems are preferred for short-term space missions, like those with a scientific objective, since they make it possible to complete them in an affordable cost. On the other hand, the resource-constrained nature of such systems poses significant challenges in software design, for the system payload. We focus on image compression and processing for scientific space missions, a vital functionality, due to the limited onboard storage capacity and the infrequent time periods, in which a satellite in orbit can send images to the ground, through a limited bandwidth connection. For CubeSat systems, the performance of image processing depends - even more than other space systems - on trade-offs, which are influenced by size, power and complexity constraints. In this context, we present the design challenges and the image compression/processing solution developed, for a university CubeSat built to carry on a biological experiment. All changes of the performance capability in the different parts of the payload image data chain (image resolution, onboard storage capacity, communication channel bandwidth and error characteristics) are taken into account. Moreover, any decision that increases the risk of incorrect reception of images from the experiment is weighted against the benefits of improved performance. We provide experimental results that show how our image processing solution resolves the associated trade-offs, while fulfilling the mission requirements.
the work is devoted to the development of methods and algorithms for forecasting of helicopters turboshaft engines technical state in flight modes based on neural network technology. Methods of probability theory and ...
详细信息
the proceedings contain 76 papers. the topics discussed include: supervised contrastive learning with multi-scale attention mechanism for fault diagnosis of bearing under variable operating conditions;dictionary learn...
ISBN:
(纸本)9781665469869
the proceedings contain 76 papers. the topics discussed include: supervised contrastive learning with multi-scale attention mechanism for fault diagnosis of bearing under variable operating conditions;dictionary learning-based intelligent recognition method towards machinery fault diagnostics;digital power system data transmission framework and configuration design;a design of a data deterministic interaction framework for improving digital twin power systems;analysis of degradation characteristics of hard disk drives in accelerated aging tests and exploration of PHM methods;reliability allocation of sonar based on fuzzy hierarchy method;determination of particle size of sodium carnallite ore in potash fertilizer production;and a cross-domain bearing fault diagnosis method towards unbalanced data based on universal domain adaptation.
In this paper, we propose a novel federated learning60;of random oblique stumps (FL-ROS) for handling the ImageNet challenge having 1,281,167 images and 1,000 classes. Our FL-ROS algorithm trains an ensemble random...
详细信息
For the hysteresis of fluid heat transfer in liquid cooling system, the detection of fluid temperature lags behind the temperature rise of heating devices; A PSO-GWO-RBF neural network temperature prediction method fo...
详细信息
ISBN:
(纸本)9781450397407
For the hysteresis of fluid heat transfer in liquid cooling system, the detection of fluid temperature lags behind the temperature rise of heating devices; A PSO-GWO-RBF neural network temperature prediction method for liquid cooling system is proposed. On the basis of RBF neural network model, PSO-GWO algorithm is used to optimize the network parameters, and the prediction model is established by using the measured parameters of the liquid cooling system. the prediction data is compared withthe real data, and the error curve is drawn. the simulation results show that the prediction effect of PSO-GWO-RBF neural network model is better than that of traditional RBF, PSO-RBF and GWO-RBF neural network models. Compared with other models, the convergence speed is faster, the convergence accuracy is higher, and the divergence is locally optimal. It has better practical value to apply the predicted temperature to the cooling control of the liquid cooling system.
Withthe continuous development of economic integration and the continuous update of electronic information technology, the status and influence of cross-border e-commerce in China's foreign trade are increasingly...
详细信息
暂无评论