期刊文献+
共找到120篇文章
< 1 2 6 >
每页显示 20 50 100
Feature-aided pose estimation approach based on variational auto-encoder structure for spacecrafts 认领 引用 被引量:2
1
作者 Yanfang LIU Rui ZHOU +2 位作者 Desong DU Shuqing CAO Naiming QI 《Chinese Journal of Aeronautics》 SCIE EI CAS CSCD 2024年第8期329-341,共13页
Real-time 6 Degree-of-Freedom(DoF)pose estimation is of paramount importance for various on-orbit tasks.Benefiting from the development of deep learning,Convolutional Neural Networks(CNNs)in feature extraction has yie... Real-time 6 Degree-of-Freedom(DoF)pose estimation is of paramount importance for various on-orbit tasks.Benefiting from the development of deep learning,Convolutional Neural Networks(CNNs)in feature extraction has yielded impressive achievements for spacecraft pose estimation.To improve the robustness and interpretability of CNNs,this paper proposes a Pose Estimation approach based on Variational Auto-Encoder structure(PE-VAE)and a Feature-Aided pose estimation approach based on Variational Auto-Encoder structure(FA-VAE),which aim to accurately estimate the 6 DoF pose of a target spacecraft.Both methods treat the pose vector as latent variables,employing an encoder-decoder network with a Variational Auto-Encoder(VAE)structure.To enhance the precision of pose estimation,PE-VAE uses the VAE structure to introduce reconstruction mechanism with the whole image.Furthermore,FA-VAE enforces feature shape constraints by exclusively reconstructing the segment of the target spacecraft with the desired shape.Comparative evaluation against leading methods on public datasets reveals similar accuracy with a threefold improvement in processing speed,showcasing the significant contribution of VAE structures to accuracy enhancement,and the additional benefit of incorporating global shape prior features. 展开更多
关键词 Pose estimation Variational auto-encoder Feature-aided Pose Estimation Approach On-orbit measurement tasks Simulated and experimental dataset
暂未订购 下载PDF
SNP site-drug association prediction algorithm based on denoising variational auto-encoder 认领 引用 被引量:2
2
作者 SONG Xiaoyu FENG Xiaobei +3 位作者 ZHU Lin LIU Tong WU Hongyang LI Yifan 《Journal of Measurement Science and Instrumentation》 CAS CSCD 2022年第3期300-308,共9页
Single nucletide polymorphism(SNP)is an important factor for the study of genetic variation in human families and animal and plant strains.Therefore,it is widely used in the study of population genetics and disease re... Single nucletide polymorphism(SNP)is an important factor for the study of genetic variation in human families and animal and plant strains.Therefore,it is widely used in the study of population genetics and disease related gene.In pharmacogenomics research,identifying the association between SNP site and drug is the key to clinical precision medication,therefore,a predictive model of SNP site and drug association based on denoising variational auto-encoder(DVAE-SVM)is proposed.Firstly,k-mer algorithm is used to construct the initial SNP site feature vector,meanwhile,MACCS molecular fingerprint is introduced to generate the feature vector of the drug module.Then,we use the DVAE to extract the effective features of the initial feature vector of the SNP site.Finally,the effective feature vector of the SNP site and the feature vector of the drug module are fused input to the support vector machines(SVM)to predict the relationship of SNP site and drug module.The results of five-fold cross-validation experiments indicate that the proposed algorithm performs better than random forest(RF)and logistic regression(LR)classification.Further experiments show that compared with the feature extraction algorithms of principal component analysis(PCA),denoising auto-encoder(DAE)and variational auto-encode(VAE),the proposed algorithm has better prediction results. 展开更多
关键词 association prediction k-mer molecular fingerprinting support vector machine(SVM) denoising variational auto-encoder(DVAE)
暂未订购 下载PDF
融合多域特征的VAE模型在肌肉疲劳分析中的应用 认领 引用
3
作者 董博 吕东澔 +1 位作者 喻大华 杜晓炜 《浙江大学学报(工学版)》 EI CAS CSCD 北大核心 2026年第6期1317-1328,共12页
针对现有肌肉疲劳分析方法中存在的特征单一、对复杂疲劳演化过程捕捉能力不足以及疲劳状态界定不清等问题,基于变分自编码器(VAE)提出结构-平滑正则化变分自编码器(SSR-VAE).结合时域、频域和非线性熵3类特征输入,增强对表面肌电(sEMG... 针对现有肌肉疲劳分析方法中存在的特征单一、对复杂疲劳演化过程捕捉能力不足以及疲劳状态界定不清等问题,基于变分自编码器(VAE)提出结构-平滑正则化变分自编码器(SSR-VAE).结合时域、频域和非线性熵3类特征输入,增强对表面肌电(sEMG)信号动态变化的捕捉能力.通过在损失函数中引入加权KL散度和轨迹平滑正则化项,优化了潜变量空间的解耦性和时序连续性.利用相关性筛选机制,从潜变量空间中提取主疲劳因子,有效表征肌肉疲劳状态.实验结果表明,与VAE模型相比,SSR-VAE在重构性能方面表现更优,利用提取的主疲劳因子,能够更清晰地划分疲劳阶段.SSR-VAE在肌肉疲劳状态分类中的最高准确率达到96.211%,明显优于其他对比方法. 展开更多
关键词 肌肉疲劳 表面肌电信号 变分自编码器(VAE) 潜变量 特征融合
暂未订购 下载PDF
基于自适应VAE的电力物联网异常流量检测 认领 引用
4
作者 张琦 龚笔华 +4 位作者 钟凯 李向明 王虎 阳跃永 彭娅莉 《邮电设计技术》 2026年第2期78-84,共7页
针对电力物联网中传统静态阈值流量异常检测方法误报率和漏报率高的问题,提出一种基于改进自适应变分自动编码器(VAE)的检测方法。该方法利用电力智能融合终端采集的流量构建流特征矩阵,设计了模型迭代与攻击检测双模块架构。网络流量... 针对电力物联网中传统静态阈值流量异常检测方法误报率和漏报率高的问题,提出一种基于改进自适应变分自动编码器(VAE)的检测方法。该方法利用电力智能融合终端采集的流量构建流特征矩阵,设计了模型迭代与攻击检测双模块架构。网络流量经攻击检测模块的初步筛选后,进入模型迭代模块进行无监督学习。模型迭代模块采用自适应阈值机制动态更新模型。在2个数据集上的实验表明,该方法有效降低了误报率和漏报率,相比传统方法有5%~8%的性能提升。 展开更多
关键词 变分自编码器 异常检测 无监督学习 电力物联网
暂未订购 下载PDF
基于VMD与LSTM-VAE密度聚类的火电机组主蒸汽压力异常检测方法 认领 引用
5
作者 廖彬生 聂益民 +3 位作者 曾江蛟 袁宇龙 何钧 袁小翠 《机械与电子》 2026年第1期103-110,共8页
为提升火电机组主控参数异常检测能力,以主控参数主蒸汽压力为对象,提出基于VMD与LSTM-VAE密度聚类的火电机组主蒸汽压力异常检测方法。采用变分模态分解方法对主蒸汽压力时序信号进行多分量分解,提取主导模态并重建信号以实现降噪;构... 为提升火电机组主控参数异常检测能力,以主控参数主蒸汽压力为对象,提出基于VMD与LSTM-VAE密度聚类的火电机组主蒸汽压力异常检测方法。采用变分模态分解方法对主蒸汽压力时序信号进行多分量分解,提取主导模态并重建信号以实现降噪;构建长短期记忆网络变分自编码器模型,通过无监督学习提取正常工况下的时序数据分布特征,计算每个滑窗的重构误差;将重构误差作为聚类特征,并应用自适应密度聚类算法对重构误差特征分类,从而实现异常检测。以某电厂650 MW火电机组主蒸汽压力数据为算例样本,分析结果表明,异常检测精确率达到0.832,召回率达到1.000,F1分数达到0.908,且在标准差0.05的高斯噪声干扰下F1分数仍达到0.887,验证了所提方法的检测性能和应用价值。 展开更多
关键词 火电机组 异常检测 变分模态分解 LSTM-VAE DBSCAN
暂未订购 下载PDF
基于时空门控VAE的ADS-B数据异常检测方法 认领 引用
6
作者 蒋东旭 刘蕾 《计算机测量与控制》 2026年第1期51-58,共8页
广播式自动相关监视是新一代空中管理系统重要的组成部分,但由于ADS-B报文以明文形式广播且缺乏数据加密和认证,导致其极易受到欺骗干扰;针对以上问题,提出一种基于时空门控变分自编码器的ADS-B数据异常检测算法;该算法编码器通过采用双... 广播式自动相关监视是新一代空中管理系统重要的组成部分,但由于ADS-B报文以明文形式广播且缺乏数据加密和认证,导致其极易受到欺骗干扰;针对以上问题,提出一种基于时空门控变分自编码器的ADS-B数据异常检测算法;该算法编码器通过采用双向LSTM建模局部时序特征,结合3层8头Transformer提取全局时空特征,并利用门控网络动态融合时空特征;引入变分推理生成潜在空间分布,约束模型对正常飞行模式的概率建模;解码器采用单层LSTM与2层Transformer的级联结构通过全连接层同步重建多维飞行参数;经实验测试,在不同攻击场景下,该模型可有效检测出ADS-B数据的各类异常,性能优于相关基线算法,为提升空中管理系统安全性提供了可行性方案。 展开更多
关键词 广播式自动相关监视(ADS-B) 异常检测 长短期记忆神经网络(LSTM) Transformer 变分自编码器(VAE)
暂未订购 下载PDF
Low Frequency Residential Load Disaggregation via Improved Variational Auto-encoder and Siamese Network 认领 引用 被引量:2
7
作者 Cheng Qian Zaijun Wu +2 位作者 Dongliang Xu Qinran Hu Yu Liu 《CSEE Journal of Power and Energy Systems》 SCIE EI CSCD 2025年第5期2137-2149,共13页
Non-intrusive load monitoring(NILM)can infer load profiles for each individual appliance from aggregated power consumption signals without installing extra sub-meters.However,performance of traditional energy disaggre... Non-intrusive load monitoring(NILM)can infer load profiles for each individual appliance from aggregated power consumption signals without installing extra sub-meters.However,performance of traditional energy disaggregation methods deteriorates in complex environments,especially susceptible to the presence of other high power consumption appliances.Practicalities are also limited by diversity of household load patterns and measurement errors.In order to address these problems,a hybrid deep learning model consisting of two steps is proposed in this paper.First,an improved variational autoencoder(VAE)structure is introduced for preliminary energy disaggregation,where the encoder and decoder layers are long short-term networks(LSTM)to extract temporal characteristics of active power signals.Afterward,a post-processing method based on Siamese one-dimensional convolutional neural network(S-1D-CNN)is adopted to remove incorrectly predicted activation segments of target appliances.Experiments are conducted on two public datasets,and results show remarkable improvements on prediction accuracy over other deep learning methods.Both transferability and stability of the proposed model are verified under different working conditions. 展开更多
关键词 Deep learning NILM post-processing Siamese network variational auto-encoder
暂未订购 下载PDF
基于VAE-BiGRU-Attention模型的储层孔隙度预测——以中-低渗砂岩储层为例 认领 引用 被引量:7
8
作者 曾滨鑫 肖晖 +1 位作者 郝子眉 刘欢欢 《地球物理学进展》 CSCD 北大核心 2025年第2期658-669,共12页
孔隙度是储层评价中不可或缺的关键物性参数,而测井曲线与其之间存在复杂且潜在的关联.以往研究中,测井曲线特征提取的不完整和模型构建较为简单,导致孔隙度预测精度受限.为提升预测精度,本文创新性地结合了变分自编码器(Variational Au... 孔隙度是储层评价中不可或缺的关键物性参数,而测井曲线与其之间存在复杂且潜在的关联.以往研究中,测井曲线特征提取的不完整和模型构建较为简单,导致孔隙度预测精度受限.为提升预测精度,本文创新性地结合了变分自编码器(Variational Auto-Encoders,VAE)、双向门控循环单元(Bidirectional Gated Recurrent Unit,BiGRU)和注意力(Attention)机制,构建了VAE-BiGRU-Attention模型.VAE能有效学习数据的潜在表示,提升数据表征能力;BiGRU擅长捕捉序列数据信息,特别适合处理孔隙度随深度变化的特征;而Attention机制的引入动态计算了每个时间步的注意力权重,从而更精准地聚焦关键特征使模型达到更好的预测效果.为验证模型的有效性,本文将其与深度神经网络(Deep Neural Network,DNN)、循环神经网络(Recurrent Neural Network,RNN)以及BiGRU-Attention进行了对比实验.结果显示,VAE-BiGRUAttention模型的均方误差(Mean Squared Error,MSE)、平均绝对误差(Mean Absolute Error,MAE)和均方根误差(Root Mean Squared Error,RMSE)分别为0.995、0.698和0.998,相较于其他模型,表现出显著的进步,有效提升了孔隙度预测的精度,为储层孔隙度预测提供了更为可靠的方法. 展开更多
关键词 孔隙度 储层评价 测井曲线 变分自编码器 双向门控循环单元 VAE-BiGRU-Attention模型
暂未订购 下载PDF
Auto-Encoding Variational Bayesian Inference in High-Dimensional Skew-Normal Linear Mixed Models 认领 引用
9
作者 YI Jieyi TANG Niansheng +1 位作者 WU Ying SU Tong 《Journal of Systems Science & Complexity》 SCIE EI CSCD 2026年第4期1763-1794,共32页
High-dimensional linear mixed models are widely used for longitudinal data analysis,yet their reliance on normality assumptions often limits applicability in psychometric and biomedical settings.To address this,the au... High-dimensional linear mixed models are widely used for longitudinal data analysis,yet their reliance on normality assumptions often limits applicability in psychometric and biomedical settings.To address this,the authors propose a high-dimensional skew-normal linear mixed model and develop a novel variational Baysian method that integrates spike-and-slab Lasso priors for simultaneous parameter estimation and variable selection.To handle dependencies in the joint posterior,the authors propose a variational auto-encoders to extract latent features,and employ a coordinate ascent algorithm to optimize the evidence lower bound(ELBO),circumventing intractable integrals.Model comparison is conducted using the Bayes factor,approximated via the ELBO.The effectiveness of the proposed methodologies is demonstrated through simulation studies and a real-data application. 展开更多
关键词 Evidence lower bound skew normal linear mixed model spike-and-slab priors variational auto-encoder variational Bayesian inference
暂未订购 下载PDF
VMGP:A unified variational auto-encoder based multi-task model for multi-phenotype,multi-environment,and cross-population genomic selection in plants 认领 引用 被引量:1
10
作者 Xiangyu Zhao Fuzhen Sun +6 位作者 Jinlong Li Dongfeng Zhang Qiusi Zhang Zhongqiang Liu Changwei Tan Hongxiang Ma Kaiyi Wang 《Artificial Intelligence in Agriculture》 SCIE EI CSCD 2025年第4期829-842,共14页
Plant breeding stands as a cornerstone for agricultural productivity and the safeguarding of food security.The advent of Genomic Selection heralds a new epoch in breeding,characterized by its capacity to harness whole... Plant breeding stands as a cornerstone for agricultural productivity and the safeguarding of food security.The advent of Genomic Selection heralds a new epoch in breeding,characterized by its capacity to harness whole-genome variation for genomic prediction.This approach transcends the need for prior knowledge of genes associated with specific traits.Nonetheless,the vast dimensionality of genomic data juxtaposed with the relatively limited number of phenotypic samples often leads to the“curse of dimensionality”,where traditional statistical,machine learning,and deep learning methods are prone to overfitting and suboptimal predictive performance.To surmount this challenge,we introduce a unified Variational auto-encoder based Multi-task Genomic Prediction model(VMGP)that integrates self-supervised genomic compression and reconstruction with multiple prediction tasks.This approach provides a robust solution,offering a formidable predictive framework that has been rigorously validated across public datasets for wheat,rice,and maize.Our model demonstrates exceptional capabilities in multi-phenotype and multi-environment genomic prediction,successfully navigating the complexities of cross-population genomic selection and underscoring its unique strengths and utility.Furthermore,by integrating VMGP with model interpretability,we can effectively triage relevant single nucleotide polymorphisms,thereby enhancing prediction performance and proposing potential cost-effective genotyping solutions.The VMGP framework,with its simplicity,stable predictive prowess,and open-source code,is exceptionally well-suited for broad dissemination within plant breeding programs.It is particularly advantageous for breeders who prioritize phenotype prediction yet may not possess extensive knowledge in deep learning or proficiency in parameter tuning. 展开更多
关键词 Genomic selection Variational auto-encoder Multi-task Deep learning Genomic prediction
Predicting the Antigenic Variant of Human Influenza A(H3N2) Virus with a Stacked Auto-Encoder Model 认领 引用
11
作者 Zhiying Tan Kenli Li +1 位作者 Taijiao Jiang Yousong Peng 《国际计算机前沿大会会议论文集》 EI 2017年第2期71-73,共3页
The influenza virus changes its antigenicity frequently due to rapid mutations, leading to immune escape and failure of vaccination. Rapid determination of the influenza antigenicity could help identify the antigenic ... The influenza virus changes its antigenicity frequently due to rapid mutations, leading to immune escape and failure of vaccination. Rapid determination of the influenza antigenicity could help identify the antigenic variants in time. Here, we built a stacked auto-encoder (SAE) model for predicting the antigenic variant of human influenza A(H3N2) viruses based on the hemagglutinin (HA) protein sequences. The model achieved an accuracy of 0.95 in five-fold cross-validations, better than the logistic regression model did. Further analysis of the model shows that most of the active nodes in the hidden layer reflected the combined contribution of multiple residues to antigenic variation. Besides, some features (residues on HA protein) in the input layer were observed to take part in multiple active nodes, such as residue 189, 145 and 156, which were also reported to mostly determine the antigenic variation of influenza A(H3N2) viruses. Overall,this work is not only useful for rapidly identifying antigenic variants in influenza prevention, but also an interesting attempt in inferring the mechanisms of biological process through analysis of SAE model, which may give some insights into interpretation of the deep learning 展开更多
关键词 Stacked auto-encoder Antigenic variation nfluenza Machine learning
暂未订购 下载PDF
基于VQ-VAE的船用设备轴承故障诊断模型 认领 引用 被引量:2
12
作者 刘建男 车驰东 《船舶工程》 CSCD 北大核心 2025年第6期53-62,共10页
[目的]针对轴承故障诊断中样本不充分、分布不均衡的问题,提出一种基于向量量化自编码器(VQ-VAE)的轴承故障诊断模型。[方法]利用VQ-VAE将轴承振动时频图压缩得到离散特征空间,并通过像素卷积神经网络(PixelCNN)采样得到全新的故障样本... [目的]针对轴承故障诊断中样本不充分、分布不均衡的问题,提出一种基于向量量化自编码器(VQ-VAE)的轴承故障诊断模型。[方法]利用VQ-VAE将轴承振动时频图压缩得到离散特征空间,并通过像素卷积神经网络(PixelCNN)采样得到全新的故障样本用于扩充、平衡轴承故障数据集。在经典轴承故障数据集进行样本生成试验,并在不同负载的轴承振动数据集上进行跨工况故障诊断迁移学习。[结果]通过生成和诊断结果的对比分析证明,提出的方法能够生成高质量的轴承故障样本对数据集进行扩充,并且能够通过迁移学习在跨工况的故障诊断中取得较高的准确率。[结论]研究结果为船用设备轴承故障诊断方法提供参考。 展开更多
关键词 故障诊断 向量量化自编码器 迁移学习 跨工况
暂未订购 下载PDF
Study of current distribution generation in PEMFC based on conditional variational auto-encoder 认领 引用
13
作者 Chengyin Shi Cong Yin +2 位作者 Weilong Luo Hailong Liu Hao Tang 《Energy and AI》 EI CSCD 2025年第3期578-591,共14页
The Proton Exchange Membrane Fuel Cell(PEMFC)converts the chemical energy of hydrogen fuel directly into electrical energy with broad application prospects.Understanding how current density is distributed in the PEMFC... The Proton Exchange Membrane Fuel Cell(PEMFC)converts the chemical energy of hydrogen fuel directly into electrical energy with broad application prospects.Understanding how current density is distributed in the PEMFC systems is crucial as it is a key factor influencing system performance.However,direct modeling for current distribution may encounter the challenge of dimensional catastrophe owing to the high dimensionality of the data.This paper uses a high-resolution segmented measurement device with 396 points to conduct experimental tests on the current distribution of a PEMFC with reactive area of 406 cm2 during a stepwise increase in load current.The current distribution is modeled based on the test results to learn the mapping relationship between the experimental parameters and the current distribution.The proposed model utilizes a Conditional Variational Auto-Encoder(CVAE)to generate current distributions.The MSE(Mean-Square Error)of the trained CVAE model reaches 9.2×10-5,and the comparison results show that the 222.9A current distribution error has the largest MSE of 6.36×10-4 and a KL Divergence(Kullback-Leibler Divergence)of 9.55×10-4,both of which are at a low level.This model enables the direct determination of the current distribution based on the experimental parameters,thereby establishing a technical foundation for investigating the impact of experimental conditions on fuel cells.This model is also of great significance for research on fuel cell system control strategies and fault diagnosis. 展开更多
关键词 Proton exchange membrane fuel cell Segmented measurement device Current distribution Conditional variational auto-encoder
Robust control strategy for semi-active air suspension systems based on reinforcement learning with entropy theory 认领 引用
14
作者 Da Wang Guoqing Zhang +3 位作者 Chunyang Qi Chuanxue Song Feng Xiao Liqiang Jin 《Acta Mechanica Sinica》 SCIE EI CAS CSCD 2026年第7期541-554,共14页
This study proposes a robust control strategy for semi-active air suspension systems(SASS)based on entropy theory.The multi-objective optimization of a system can be described as a long-term problem using entropy valu... This study proposes a robust control strategy for semi-active air suspension systems(SASS)based on entropy theory.The multi-objective optimization of a system can be described as a long-term problem using entropy values by innovatively introducing entropy theory.The state marginal probability of the SASS is incorporated into the reward function as the entropy value.This incorporation incentivizes the agent to focus on reducing the entropy value of the system state over a period of time during the exploration process,thereby reducing the degree of coupling between system states.This study also proposes an optimization strategy that introduces a state observer based on a variational auto-encoder.The observer can extract environmental features from historical states and expand the dimension of the state,thereby enhancing the generalization performance of the system under different road excitations.Bench test results show that the algorithm improves ride comfort while ensuring robustness.The root mean square(RMS)of body vertical acceleration decreased by 13.01%,while the RMS of dynamic tyre displacement only increased by 2.36%. 展开更多
关键词 Semi-active air suspension Entropy theory Robust controls Variational auto-encoder
暂未订购 下载PDF
电动汽车锂离子电池预测-评估故障检测框架 认领 引用 被引量:1
15
作者 廖雪超 陈睿 《计算机应用》 CSCD 北大核心 2026年第5期1614-1623,共10页
针对电动汽车锂离子电池故障检测中多源异构时序数据复杂性高、异常样本稀缺及多变量关联性强的挑战,提出一种基于动态变换记忆自编码器的预测-评估故障检测框架(DTMAD),以提升故障识别准确性与模型泛化能力。首先,设计融合动态自编码器... 针对电动汽车锂离子电池故障检测中多源异构时序数据复杂性高、异常样本稀缺及多变量关联性强的挑战,提出一种基于动态变换记忆自编码器的预测-评估故障检测框架(DTMAD),以提升故障识别准确性与模型泛化能力。首先,设计融合动态自编码器(DyAD)与门控循环单元(GRU)的联合特征编码器,对多源时序数据进行特征融合与降维处理,提取跨模态深层特征表示;同时,构建基于自注意力机制的预响应编码器,捕捉时序数据中的长期依赖关系,提升特征提取效率与精度;进一步地,引入记忆解析模块,通过残差对比学习机制融合预测路径与实际响应路径,增强模型对异常模式的检测能力。其次,基于重构误差的分布特性,通过协同异常检测算法设计评估模型。最后,通过综合的预测-评估框架,在无监督学习条件下从多源数据中提取关键响应模式并识别潜在异常。在多组多源电动汽车锂离子电池数据集上的实验结果表明,所提框架的故障检测准确率和模型稳定性均优于对比的编码器(AE)、深度支持向量数据描述(DeepSVDD)与图偏差网络(GDN)等。其中,相较于DyAD模型,DTMAD模型的接收者操作特征曲线下面积(AUROC)提升至0.9008,且结果波动幅度由0.029降至0.026,展现出更高的检测稳定性与泛化能力。 展开更多
关键词 锂离子电池 故障检测 特征融合 变分自编码器
暂未订购 下载PDF
基于TCN-VAE-注意力机制的超高层建筑多模态能耗数据处理与自适应预测模型研究 认领 引用
16
作者 刘冠显 《软件》 2025年第12期46-48,共3页
针对超高层建筑多模态能耗数据采集传输中的传感器异常、时序错乱、噪声干扰问题导致传统模型预测误差超15%的痛点,本文提出了融合时序卷积网络(TCN)、变分自编码器(VAE)与注意力机制的多模态数据处理与自适应预测模型。模型通过“噪声... 针对超高层建筑多模态能耗数据采集传输中的传感器异常、时序错乱、噪声干扰问题导致传统模型预测误差超15%的痛点,本文提出了融合时序卷积网络(TCN)、变分自编码器(VAE)与注意力机制的多模态数据处理与自适应预测模型。模型通过“噪声检测—特征提纯—动态预测”三级架构,实现了多源数据降噪与1小时/12小时/24小时多尺度能耗预测协同优化。在超高层写字楼、酒店、商业体三类场景仿真数据集上的实验结果表明,1小时预测MAPE(平均绝对百分比误差)低至7.8%,较传统LSTM降低42.6%;15%高噪声下MAPE仅10.2%,鲁棒性显著优于对比模型;工程仿真中空调日均能耗降低9.7%,电网负荷波动缩小12.3%。本文研究为超高层建筑能耗预测提供了数据驱动的计算机技术方案,为实地部署奠定了算法基础。 展开更多
关键词 超高层建筑 多模态数据处理 噪声抑制 时序卷积网络(TCN) 变分自编码器(VAE) 注意力机制 能耗预测
暂未订购 下载PDF
基于时空上下文感知的解纠缠兴趣点推荐 认领 引用
17
作者 杨晓文 李锦翔 +3 位作者 况立群 孙福盛 庞敏 李潞洋 《计算机应用研究》 CSCD 北大核心 2026年第3期858-866,共9页
现有PoI推荐方法在时空上下文建模与兴趣解纠缠方面存在不足,难以兼顾多层次的时空依赖,且用户兴趣容易混淆,限制了对多样性兴趣和冷门PoI的发现。针对上述问题,提出了一种基于时空上下文感知的解纠缠兴趣点推荐模型(ST-DPR)。该模型设... 现有PoI推荐方法在时空上下文建模与兴趣解纠缠方面存在不足,难以兼顾多层次的时空依赖,且用户兴趣容易混淆,限制了对多样性兴趣和冷门PoI的发现。针对上述问题,提出了一种基于时空上下文感知的解纠缠兴趣点推荐模型(ST-DPR)。该模型设计了基于Transformer结构的变分自编码器模块(DIDVAE),用两个独立编码器分别建模主要兴趣与多样性兴趣,以刻画不同兴趣模式之间的差异性。利用分层编码器进一步捕捉签到序列中的局部与全局时空上下文信息,实现对用户偏好的精细建模。训练阶段结合交叉熵损失、VAE的重构损失、KL散度和互信息损失以提升预测性能并促进兴趣解纠缠。基于Foursquare NYC、TKY和US三个真实数据集的实验表明,ST-DPR在命中率(HR)和归一化折损累积增益(NDCG)指标上优于现有先进模型,验证了其在PoI预测任务中的有效性与优越性。 展开更多
关键词 兴趣点推荐 时空上下文 变分自编码器 解纠缠 互信息
暂未订购 下载PDF
端到端智能图像视频编码的发展回顾与前沿展望 认领 引用
18
作者 陈彤 陆明 +11 位作者 石峻奇 丛吾洋 丁丹丹 贾川民 刘家瑛 刘东 宋利 马思伟 杨铀 刘文予 曹汛 马展 《中国图象图形学报》 CSCD 北大核心 2026年第6期1595-1618,共24页
图像与视频编码及相应标准自诞生以来,一直支撑着点播、直播以及视频会议等核心多媒体服务。过去30余年,主流技术路线围绕规则驱动的模块化工具(如变换、预测、熵编码、环路滤波等)的精细化设计与协同优化展开,并借助标准化组织形成生... 图像与视频编码及相应标准自诞生以来,一直支撑着点播、直播以及视频会议等核心多媒体服务。过去30余年,主流技术路线围绕规则驱动的模块化工具(如变换、预测、熵编码、环路滤波等)的精细化设计与协同优化展开,并借助标准化组织形成生态。近10年,随着深度学习表征能力、公共数据集累积以及高效训练/推理框架的成熟,端到端智能编码技术快速迭代,在若干测试集与应用场景中展现出超越传统标准的压缩性能。本报告围绕图像编码,第1部分概述端到端智能编码主流框架演化主线;第2部分阐述率失真性能指标之外的可实用性功能,包括可变码率与码率控制、模型量化与鲁棒性;第3部分总结智能编码纳入/影响标准化进程的努力与现状;第4部分探讨从智能图像编码到智能视频编码的进一步拓展。希望本文能够为研究者与工程实践者提供系统化的思考视角,促进智能图像视频编码方法在产业级场景中的有序落地。 展开更多
关键词 智能图像压缩 变分自编码器(VAE) 率失真(R-D) 实用性 标准化
暂未订购 下载PDF
基于超球环描述的概率性结构损伤识别 认领 引用
19
作者 郭茂祖 张庆宇 +1 位作者 赵玲玲 邓扬 《计算机应用》 CSCD 北大核心 2026年第6期2016-2025,共10页
土木工程结构损伤识别中无监督的阈值方法无需标注数据,然而数据的不确定性导致阈值附近存在识别不准的问题。针对无监督结构损伤识别阈值方法在阈值附近存在误报和漏报的问题,基于深度支持向量数据描述(Deep-SVDD)提出基于超球环描述... 土木工程结构损伤识别中无监督的阈值方法无需标注数据,然而数据的不确定性导致阈值附近存在识别不准的问题。针对无监督结构损伤识别阈值方法在阈值附近存在误报和漏报的问题,基于深度支持向量数据描述(Deep-SVDD)提出基于超球环描述的概率性结构损伤识别方法VAEKL-RDDP(Variational AutoEncoder with Kullback-Leibler divergence constrained for hypersphere Ring Data Description Probabilistic damage identification)。该方法以变分自编码器(VAE)为框架,利用KL(Kullback-Leibler)散度约束构造超球环。首先,预训练VAE,以重建结构加速度响应;其次,引入KL散度,以联合训练预训练的VAE编码器与超球环描述方法,从而在加速度数据特征的后验分布中提取可靠分类边界;最后,依据分类边界构造超球环,依托所构造的超球环对结构进行损伤识别,并采用累积概率密度方法评估超球环内数据。在真实的Z24桥结构的渐进性损伤和足尺木亭振动台实验中,与基于自编码器(AE)重建的基线方法相比,VAEKL-RDDP的准确率和召回率分别平均提高了24.9%和36.7%;而相较于Deep-SVDD和扩散模型的插补预测(ImDiffusion)等方法,VAEKL-RDDP的准确率和召回率分别平均提升了20.8%和33.7%,验证了所提方法提高了损伤检测的性能,降低了漏报可能性。 展开更多
关键词 无监督结构损伤识别 时间序列重构 概率性损伤评估 累积概率密度 变分自编码器
暂未订购 下载PDF
基于增强变分自编码器的锂离子电池健康状态估计 认领 引用 被引量:1
20
作者 孙剑 马朋举 +4 位作者 陈鑫 刘彦铄 陈明皓 杨浩坤 王凯 《广东电力》 北大核心 2026年第1期20-33,共14页
锂离子电池的阻抗作为重要的电化学信息,能够反映电池健康状态。现实测试中的阻抗数据常受到噪声影响,给健康状态估计带来挑战,为此提出一种适用于噪声数据的电池健康状态估计模型。首先,将变分模态分解和变分自编码器融合,提出一种新... 锂离子电池的阻抗作为重要的电化学信息,能够反映电池健康状态。现实测试中的阻抗数据常受到噪声影响,给健康状态估计带来挑战,为此提出一种适用于噪声数据的电池健康状态估计模型。首先,将变分模态分解和变分自编码器融合,提出一种新的特征提取模型,称为增强变分自编码器(enhanced variational autoencoder,EVAE),EVAE对不同工况的阻抗数据进行编码并提取强相关性特征;在此基础上,使用Transformer模型将所提特征与健康状态进行拟合;最后,通过在商用18650型电池上的实验,验证该模型的鲁棒性。实验结果表明,所提模型能够在不同工况的阻抗数据下有效估计锂离子电池的健康状态,可显著提高特征提取和健康状态估计的性能。该算法可有效解决现实测试中噪声对阻抗数据的影响,准确估计健康状态,有助于提高电池的寿命和性能。 展开更多
关键词 锂离子电池 健康状态 变分模态分解 变分自编码器 增强变分自编码器 Transformer模型
暂未订购 下载PDF
上一页 1 2 6 下一页 到第
在线咨询 使用帮助 返回顶部 意见反馈