Loss functions are essential for optimizing the imaging performance of trained deep learning models. Thus, the design of tailored loss functions defines the effectiveness of deep learning imaging models. A loss functi...
详细信息
The balance of supply and demand is pivotal in ensuring efficient and reliable power grid utilization. With the growth of demand, the integration of renewable energy resources (RERs) into the power grid is increasing ...
详细信息
Sleep staging serves as a fundamental assessment for sleep quality measurement and sleep disorder diagnosis. Although current deep learning approaches have successfully integrated multimodal sleep signals, enhancing t...
详细信息
Sleep staging serves as a fundamental assessment for sleep quality measurement and sleep disorder diagnosis. Although current deep learning approaches have successfully integrated multimodal sleep signals, enhancing the accuracy of automatic sleep staging, certain challenges remain, as follows: 1) optimizing the utilization of multi-modal information complementarity, 2) effectively extracting both long- and short-range temporal features of sleep information, and 3) addressing the class imbalance problem in sleep data. To address these challenges, this paper proposes a two-stream encode-decoder network, named TSEDSleepNet, which is inspired by the depth sensitive attention and automatic multi-modal fusion (DSA2F) framework. In TSEDSleepNet, a two-stream encoder is used to extract the multiscale features of electrooculogram (EOG) and electroencephalogram (EEG) signals. And a self-attention mechanism is utilized to fuse the multiscale features, generating multi-modal saliency features. Subsequently, the coarser-scale construction module (CSCM) is adopted to extract and construct multi-resolution features from the multiscale features and the salient features. Thereafter, a Transformer module is applied to capture both long- and short-range temporal features from the multi-resolution features. Finally, the long- and short-range temporal features are restored with low-layer details and mapped to the predicted classification results. Additionally, the Lovász loss function is applied to alleviate the class imbalance problem in sleep datasets. Our proposed method was tested on the Sleep-EDF-39 and Sleep-EDF-153 datasets, and it achieved classification accuracies of 88.9% and 85.2% and Macro-F1 scores of 84.8% and 79.7%, respectively, thus outperforming conventional traditional baseline models. These results highlight the efficacy of the proposed method in fusing multi-modal information. This method has potential for application as an adjunct tool for diagnosing sleep disorde
Frequency stability is crucial for the proper operation of the power grids. The changing landscapes in power systems, raised by continuously increasing demand and rapid decommissioning of conventional generation, have...
详细信息
This paper introduces a novel half-bridge (HB)/dual-stacked-switches based electrolytic capacitor-less bidirectional AC/DC converter for high voltage (HV) electric vehicle (EV) systems. The proposed two-stage AC/DC co...
详细信息
This study uses quantum-inspired techniques to ad-dress the DC optimal power flow problem considering frequency constraints. Although numerous analytical and data-driven meth-ods have been developed to solve DC-OPF un...
详细信息
With the rise of cloud computing, multi-user scenarios have become a common setting for data sharing nowadays. The conservative security notion might not be sufficient for such a data sharing model. As a response to t...
详细信息
Optimizing the charging protocol for large-scale electric vehicles is complex and computationally costly. Therefore, this paper proposes an advanced approach using machine learning-assisted mean field game theory to h...
详细信息
In sub nanometer carbon nanotubes,water exhibits unique dynamic characteristics,and in the high-frequency region of the infrared spectrum,where the stretching vibrations of the internal oxygen-hydrogen(O-H)bonds are c...
详细信息
In sub nanometer carbon nanotubes,water exhibits unique dynamic characteristics,and in the high-frequency region of the infrared spectrum,where the stretching vibrations of the internal oxygen-hydrogen(O-H)bonds are closely related to the hydrogen bonds(H-bonds)network between water ***,it is crucial to analyze the relationship between these two *** this paper,the infrared spectrum and motion characteristics of the stretching vibrations of the O-H bonds in one-dimensional confined water(1DCW)and bulk water(BW)in(6,6)single-walled carbon nanotubes(SWNT)are studied by molecular dynamics *** results show that the stretching vibrations of the two O-H bonds in 1DCW exhibit different frequencies in the infrared spectrum,while the O-H bonds in BW display two identical main frequency *** analysis using the spring oscillator model reveals that the difference in the stretching amplitude of the O-H bonds is the main factor causing the change in vibration frequency,where an increase in stretching amplitude leads to a decrease in spring stiffness and,consequently,a lower vibration frequency.A more in-depth study found that the interaction of H-bonds between water molecules is the fundamental cause of the increased stretching amplitude and decreased vibration frequency of the O-H ***,by analyzing the motion trajectory of the H atoms,the dynamic differences between 1DCW and BW are clearly *** findings provide a new perspective for understanding the behavior of water molecules at the nanoscale and are of significant importance in advancing the development of infrared spectroscopy detection technology.
A dandelion algorithm(DA) is a recently developed intelligent optimization algorithm for function optimization problems. Many of its parameters need to be set by experience in DA,which might not be appropriate for all...
详细信息
A dandelion algorithm(DA) is a recently developed intelligent optimization algorithm for function optimization problems. Many of its parameters need to be set by experience in DA,which might not be appropriate for all optimization problems. A self-adapting and efficient dandelion algorithm is proposed in this work to lower the number of DA's parameters and simplify DA's structure. Only the normal sowing operator is retained;while the other operators are discarded. An adaptive seeding radius strategy is designed for the core dandelion. The results show that the proposed algorithm achieves better performance on the standard test functions with less time consumption than its competitive peers. In addition, the proposed algorithm is applied to feature selection for credit card fraud detection(CCFD), and the results indicate that it can obtain higher classification and detection performance than the-state-of-the-art methods.
暂无评论