Results 1 to 10 of about 844 (118)
全球能源互联网, 2023 随着新能源并网比例持续上升，电池储能系统（battery energy storage system，BESS）的发展备受关注。逐渐完善其激励政策和市场机制，提升其经济效益，对其未来发展具有重要意义。但目前BESS的不同物理特性、政策和市场机制、退役电池的处理方式等问题仍亟待深入研究。分析了BESS的研究现状，包括不同物理特性、参与电力系统价值评估现状。从国内外储能政策的提出、电力市场的运营、碳市场的交易等方面综述了国内外BESS需求及市场机制，并分析了梯次利用BESS的适用场景与经济效益 ...
索克兰, 程林*, 许鹤麟, 黄文瑞
doaj +1 more source
Zhejiang Daxue xuebao. Lixue ban, 2022 由于在包含表情的视频数据集中存在大量与表情特征无关的视频帧，使得模型在训练中学习到大量无关信息，导致识别率大幅下降，因此如何令模型自主地选择视频关键帧成为研究的关键。在已有的视频表情识别方法中，大多没有考虑关键帧和非关键帧对模型训练效果的影响，为此提出了一种基于注意力机制与GhostNet的人脸表情识别（FSAGN）模型。通过自注意力机制与帧选择损失计算不同帧的权重，根据权重自主选择视频序列的关键帧。此外，为减少模型参数、降低模型的训练成本，将传统的特征提取网络替换为训练参数较少的GhostNet网络 ...
ZHUJintai(祝锦泰) +4 more
doaj +1 more source
智能科学与技术学报, 2023 针对肺部细胞病理图像亮度不均衡、异常细胞轮廓精准分割难以实现的问题，提出一种以U-Net为基本框架，结合稠密块以及注意力机制的异常细胞分割模型。首先，利用具有编码器-解码器结构的U-Net对异常细胞进行分割；然后，在U-Net中引入稠密块，以提高特征之间的传播能力，提取更多异常细胞的特征信息；最后，利用注意力机制提高异常细胞区域的权重，降低亮度不均衡对模型的干扰。实验结果表明，该方法的IoU和Dice相似系数值分别为0.6928和0.8060，与其他模型相比 ...
崔文成, 王可丽, 邵虹
doaj +1 more source
Research on berth occupancy prediction model based on attention mechanism [PDF]
, 2020 To solve the problem that the berth occupancy prediction accuracy decreases while the prediction step was increasing, a berth occupancy prediction model based on an attention mechanism was proposed, which was the multivariate time pattern information ...
Qindong SUN +5 more
core +1 more source
Modulation recognition algorithm based on mixed attention prototype network
Xibei Gongye Daxue Xuebao, 2022 针对极少量带标签样本条件下的通信信号调制识别难题, 提出一种基于混合注意力原型网络的调制识别算法。综合元学习和度量学习的思想, 在原型网络框架下通过特征提取模块将信号映射至一个新的特征度量空间, 并通过比较该空间内各类原型与查询信号之间的距离确定查询信号调制样式。根据通信信号IQ分量的时序特点设计了由卷积神经网络和长短时记忆网络级联的特征提取模块, 并引入卷积注意力机制提升关键特征的权重; 采用基于Episode的训练策略, 使算法可泛化到新的信号识别任务中。仿真结果表明 ...
PANG Yiqiong +4 more
doaj +1 more source
Person Re-identification Method Based on GoogLeNet-GMP Based on Vector Attention Mechanism [PDF]
, 2022 In order to improve the accuracy and applicability of person re-identification(Re-ID),a Re-ID method based on vector attention mechanism GoogLeNet is proposed.Firstly,three groups of images(anchor,positive and negative) are input into the GoogLeNet-GMP ...
MENG Yue-bo, MU Si-rong, LIU Guang-hui, XU Sheng-jun, HAN Jiu-qiang
core +1 more source
Short-term power load forecasting method based on CNN-SAEDN-Res
, 2023 In deep learning, the load data with non-temporal factors are difficult to process by sequence models. This problem results in insufficient precision of the prediction. Therefore, a short-term load forecasting method based on convolutional neural network
Cui, Yang +4 more
core +1 more source
Combined flow prediction model for natural gas pipeline network based on EMD-Attention-GRU [PDF]
, 2023 In order to overcome the deficiency of the traditional time series prediction method in flow prediction of natural gas pipeline networks, a combined prediction model based on Empirical Mode Decomposition (EMD), Attention and Gated Recurrent Unit (GRU ...
Hongxian LIN +4 more
core +1 more source
Zhejiang Daxue xuebao. Lixue ban, 2023 乳腺癌是全球最常见的恶性肿瘤之一，采用传统方法诊断需花费大量时间和精力，且受个人能力影响较大。用计算机辅助诊断的方法，可以提高病理图像分类的准确率和效率，从而满足临床应用的需求。为此，提出一种基于DenseNet的融合多尺度特征和注意力机制的乳腺癌病理图像分类算法（MFDC-Net）。在密集块中引入坐标注意力机制，精准定位重要特征的空间信息。采用多尺度池化过渡层，通过不同卷积核的平均池化和普通卷积，在实现降维的同时扩大感受野。采用多尺度特征增强模块，融合深层次图像特征，提高分类性能。结果显示，MFDC ...
方于华(FANG Yuhua) +1 more
doaj +1 more source
Remote Sensing Change Detection Based on Feature Fusion and Attention Network [PDF]
, 2022 Change detection is one of the essential tasks in remote sensing,which is usually regarded as a pixel-level classification problem.In recent years,deep neural networks have also been widely used in the change detection task due to their powerful ...
LAN Ling-xiang, CHI Ming-min
core +1 more source