检索结果-南通市图书馆

Dynamic prediction of moving trajectory in pipe jacking: gru-based deep learning framework

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Structural and Civil Engineering 2023年第7期17卷 994-1010页

作者： Yi-Feng YANG Shao-Ming LIAO Meng-Bo LIU Department of Geotechnical Engineering College of Civil EngineeringTongji UniversityShanghai 200092China Key Laboratory of Road and Traffic Engineering of Ministry of Education Tongji UniversityShanghai 201804China

The moving trajectory of the pipe-jacking machine(PJM),which primarily determines the end quality of jacked tunnels,must be controlled strictly during the entire jacking *** prediction models to support drivers in performing rectifications in advance can effectively avoid considerable trajectory deviations from the designed jacking ***,a gated recurrent unit(gru)-based deep learning framework is proposed herein to dynamically predict the moving trajectory of the *** this framework,operational data are first extracted from a data acquisition system;subsequently,they are preprocessed and used to establish gru-based multivariate multistep-ahead direct prediction *** verify the performance of the proposed framework,a case study of a large pipe-jacking project in Shanghai and comparisons with other conventional models(i.e.,long short-term memory(LSTM)network and recurrent neural network(RNN))are *** addition,the effects of the activation function and input time-step length on the prediction performance of the proposed framework are investigated and *** results show that the proposed framework can dynamically and precisely predict the PJM moving trajectory during the pipe-jacking process,with a minimum mean absolute error and root mean squared error(RMSE)of 0.1904 and 0.5011 mm,*** RMSE of the gru-based models is lower than those of the LSTM-and RNN-based models by 21.46%and 46.40%at the maximum,*** proposed framework is expected to provide an effective decision support for moving trajectory control and serve as a foundation for the application of deep learning in the automatic control of pipe jacking.

关键词： dynamic prediction moving trajectory pipe jacking gru deep learning

RogruT: A Hybrid Deep Learning Model for Detecting Power Trapping in Smart Grids

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第5期79卷 3175-3192页

作者： Farah Mohammad Saad Al-Ahmadi Jalal Al-Muhtadi Center of Excellence in Information Assurance(CoEIA) King Saud UniversityRiyadh11543Saudi Arabia College of Computer&Information Sciences King Saud UniversityRiyadh11543Saudi Arabia

Electricity theft is a widespread non-technical issue that has a negative impact on both power grids and electricity *** hinders the economic growth of utility companies,poses electrical risks,and impacts the high energy costs borne by *** development of smart grids is crucial for the identification of power theft since these systems create enormous amounts of data,including information on client consumption,which may be used to identify electricity theft using machine learning and deep learning ***,there also exist different solutions such as hardware-based solutions to detect electricity theft that may require human resources and expensive ***-based solutions are presented in the literature to identify electricity theft but due to the dimensionality curse,class imbalance issue and improper hyper-parameter tuning of such models lead to poor *** this research,a hybrid deep learning model abbreviated as RogruT is proposed to detect electricity theft as amalicious and non-malicious *** key steps of the RogruT are data preprocessing that covers the problem of class imbalance,feature extraction and final theft *** advanced-level models like RoBERTa is used to address the curse of dimensionality issue,the near miss for class imbalance,and transfer learning for *** effectiveness of the RogruTis evaluated using the dataset fromactual smartmeters.A significant number of simulations demonstrate that,when compared to its competitors,the RogruT achieves the best classification *** performance evaluation of the proposed model revealed exemplary results across *** accuracy achieved was 88%,with precision at an impressive 86%and recall reaching 84%.The F1-Score,a measure of overall performance,stood at 85%.Furthermore,themodel exhibited a noteworthyMatthew correlation coefficient of 78%and excelled with an area under the curve of 91%.

关键词： Electricity theft smart grid RoBERTa gru transfer learning

Predicting Resource Availability in Local Mobile Crowd Computing Using Convolutional gru

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第3期70卷 5199-5212页

作者： Pijush Kanti Dutta Pramanik Nilanjan Sinhababu Anand Nayyar Mehedi Masud Prasenjit Choudhury Department of Computer Science&Engineering National Institute of TechnologyDurgapurIndia Reliability Engineering Centre Indian Institute of Technology KharagpurIndia Graduate School Duy Tan UniversityDa Nang550000Vietnam Faculty of Information Technology Duy Tan UniversityDa Nang550000Vietnam Department of Computer Science College of Computers and Information TechnologyTaif UniversityTaif21944Saudi Arabia

In mobile crowd computing(MCC),people’s smart mobile devices(SMDs)are utilized as computing *** the ever-growing computing capabilities of today’s SMDs,a collection of them can offer significantly high-performance computing *** a localMCC,the SMDs are typically connected to a local Wi-Fi *** and institutions can leverage the SMDs available within the campus to form local MCCs to cater to their computing needs without any financial and operational *** it offers an economical and sustainable computing solution,users’mobility poses a serious issue in the QoS of *** address this,before submitting a job to an SMD,we suggest estimating that particular SMD’s availability in the network until the job is *** this,we propose a convolutional gru-based prediction model to assess how long an SMD is likely to be available in the network from any given point of *** experimental purposes,we collected real users’mobility data(in-time and outtime)with respect to a Wi-Fi access *** build the prediction model,we presented a novel feature extraction method to be applied to the time-series *** experimental results prove that the proposed convolutional gru model outperforms the conventional gru model.

关键词： Resource selection resource availability mobile grid mobile cloud ad-hoc cloud crowd computing deep learning gru CNN RNN

Detection method of absence seizures based on Resnet and bidirectional gru

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Acta Epileptologica 2023年第1期6卷 34-42页

作者： Lijun Li Hengxing Zhang Xiaomei Liu Jie Li Lei Li Dan Liu Jieqing Min Ping Zhu Huan Xia Shangkun Wang Li Wang Kunming children‘s hospital Kunming 650000China Zhengzhou Zoneyet Technology Corp. LtdZhengzhou 450000China

Background Epilepsy is a common chronic neurological *** repeated seizure attacks have a great negative impact on patients’physical and mental *** diagnosis of epilepsy mainly depends on electroencephalogram(EEG)signals detection and *** are two main EEG signals detection methods for *** is the detection based on abnormal waveform,the other is the analysis of EEG signals based on the traditional machine *** feature extraction method of the traditional machine learning is difcult to capture the high-dimension information between adjacent *** In this paper,redundant information was removed from the data by Gaussian fltering,downsampling,and short-time Fourier *** Neural Networks(CNN)was used to extract the high-dimensional features of the preprocessed data,and then Gate Recurrent Unit(gru)was used to combine the sequence information before and after,to fully integrate the adjacent information EEG signals and improve the accuracy of the model *** Four models were designed and *** experimental results showed that the prediction model based on deep residual network and bidirectional gru had the best efect,and the test accuracy of the absence epilepsy test set reached 92%.Conclusions The prediction time of the network is only 10 sec when predicting four-hour EEG *** can be efectively used in EEG software to provide reference for doctors in EEG analysis and save doctors’time,which has great practical value.

关键词： Resnet Epilepsy gru Electroencephalogram

基于gru-GCN 模型的碳排放权价格预测及影响因素分析研究

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

基于ABC-LSTM-gru的时间序列分解与预测模型

运筹与模糊学 2024年第2期14卷 1156页

作者：白鑫怡崔骥南京信息工程大学数学与统计学院，江苏南京苏州博纳讯动软件有限公司，江苏苏州

中国作为世界第一大碳排放国，先后成立了深圳、北京、上海等8个碳排放权交易试点，相较于国际上成熟的碳市场，中国的碳排放权交易市场尚处在探索阶段，研究全国碳市场交易价格的波动特征，有助于企业规避交易价格风险、制定绿色创新... 详细信息

中国作为世界第一大碳排放国，先后成立了深圳、北京、上海等8个碳排放权交易试点，相较于国际上成熟的碳市场，中国的碳排放权交易市场尚处在探索阶段，研究全国碳市场交易价格的波动特征，有助于企业规避交易价格风险、制定绿色创新及节能减排的战略，也对政府完善碳排放权交易市场机制、建设全国碳交易市场具有一定意义，本文以上海交易所为例，探究不同因素对碳排放权交易价格的影响程度，并且使用StemGNN模型，即基于gru和GCN具有联合捕获序列间相关性和时间依赖性功能的新型预测模型对碳排放权交易价格进行预测，以期为我国碳交易市场的参与者及政府相关管理部门提供更加科学、精准的决策工具。

关键词：碳排放权交易深度学习 gru GCN Carbon Emission Trading Deep Learning gru GCN

来源：博看期刊评论

在线全文

博看期刊

学校读者我要写书评

暂无评论

软件工程 2024年第3期27卷 58-62页

作者：朱子敬何利文南京邮电大学物联网学院江苏南京210003

针对金融时间序列数据的高噪声、时间依赖性等问题,提出了一种人工蜂群算法-长短期记忆-门控单元(ABC-LSTM-gru)混合模型。该模型综合利用长短期记忆网络(LSTM)和门控循环单元(gru)循环神经网络,更全面地捕捉时间序列中的长期和短期关... 详细信息

针对金融时间序列数据的高噪声、时间依赖性等问题,提出了一种人工蜂群算法-长短期记忆-门控单元(ABC-LSTM-gru)混合模型。该模型综合利用长短期记忆网络(LSTM)和门控循环单元(gru)循环神经网络,更全面地捕捉时间序列中的长期和短期关系。在特征处理阶段,通过相关性分析对特征进行筛选,同时采用奇异谱分析(SSA)对数据进行分解,得到高频、中频和低频三个部分。在模型的超参数优化中,采用了改进后的人工蜂群算法(ABC),以提高模型的性能。为验证ABC-LSTM-gru混合模型的有效性,选择NIFTY-50股票指数进行实证分析。实验结果对比显示,ABC-LSTM-gru混合模型在时间序列预测方面的表现更佳,相较于LSTM与gru模型,其在均方根误差(RMSE)指标上分别降低了28.3%与21.5%,显示出更为准确的预测性能。

关键词： gru LSTM ABC SSA 股市预测

一种基于改进CNN-gru的建筑冷负荷单步预测方法

维普期刊数据库

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

机电工程技术 2024年第1期53卷 119-122页

作者：赖添城徐康康朱成就蔡建阳广东工业大学机电工程学院广州510006

在智能建筑环境中,准确的建筑负荷需求预测对建筑能源管理至关重要。这在拟定高效的需求调度策略方面担任着重要角色,它提高了建筑能源利用效率,挖掘了建筑节能潜力。将卷积神经网络(CNN)和门控循环单元(gru)作为基础模型,提出了一个混... 详细信息

在智能建筑环境中,准确的建筑负荷需求预测对建筑能源管理至关重要。这在拟定高效的需求调度策略方面担任着重要角色,它提高了建筑能源利用效率,挖掘了建筑节能潜力。将卷积神经网络(CNN)和门控循环单元(gru)作为基础模型,提出了一个混合深度神经网络模型,并使用批标准化和Dropout进行改进,以准确预测公共建筑的未来1 h冷负荷需求。所提出的模型有着良好的短期记忆能力,可以高效学习数据的时间依赖性,并充分利用少量的输入特征以获得高预测精度。所提出的模型在实际建筑中进行了实验验证。结果表明,应用批标准化和Dropout有助于CNN-gru更有效地学习。从这个角度上看,冷负荷预测的RMSE值从310.02降低到261.87,MAE值从230.83降低到206.23,MAPE值从9.68%降低到8.51%,R2值从0.76提高到0.83。与其他基准模型相比,改进CNN-gru获得了最好的预测性能。

关键词： CNN gru 多维时间序列建筑冷负荷

基于FISVDD与gru的变压器声纹识别技术

维普期刊数据库

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

高电压技术 2022年第11期48卷 4546-4556页

作者：王荣昊李喆孙正胡赵宇孙汉文江秀臣上海交通大学电气工程系上海200240 上海汽车集团股份有限公司上海200041

为了降低声纹识别算法在低信噪比环境下的误报率,提出了一种基于快速增量式支持向量数据描述(fast incremental support vector data description,FISVDD)以及门控循环单元(gate recurrent unit,gru)的变压器机械故障声纹识别方法。以... 详细信息

为了降低声纹识别算法在低信噪比环境下的误报率,提出了一种基于快速增量式支持向量数据描述(fast incremental support vector data description,FISVDD)以及门控循环单元(gate recurrent unit,gru)的变压器机械故障声纹识别方法。以变压器为实验对象,分别获取变压器在正常工况、铁芯松动、线圈松动3种状态下的声音信号,并使用Mel频率倒谱系数进行特征提取。FISVDD作为第1级算法分离陌生类,同时通过增量学习的方式学习新样本。gru作为第2级分类算法,对通过第1级算法的样本进行识别。实验结果表明:与传统闭集识别算法相比,FISVDD需要的训练时间更少;相较于传统的机器学习算法,gru在变压器音频识别任务中具有更高的识别准确率和抗噪识别能力;所提方法相较于1级算法更加有效,在识别信噪比高于10 dB的样本时召回率下降不超过1%,在识别信噪比不超过0 dB的样本时误报率不超过10%。

关键词：变压器机械故障诊断增量学习 gru FISVDD

基于UNet+gru深度学习方法提高叠后地震资料分辨率

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Applied Geophysics 2023年第2期20卷 176-185,241页

作者：郭爱华路鹏飞王丹丹吴吉忠陈晓彭怀宇蒋书豪东华理工大学信息工程学院江西南昌330013 华北油田第一采油厂地质研究所河北任丘062550 东北石油大学河北秦皇岛066000 东华理工大学地球物理与测控技术学院江西南昌330013

现有的地震资料提频方法大部分都有各自的局限性,根据已有方法的优缺点,本文尝试将深度学习技术应用到提高地震资料分辨率中。首先,在UNet深度学习方法的基础上,井震结合,以测井声波时差和密度曲线建立合成地震记录,以井上合成地震记录... 详细信息

现有的地震资料提频方法大部分都有各自的局限性,根据已有方法的优缺点,本文尝试将深度学习技术应用到提高地震资料分辨率中。首先,在UNet深度学习方法的基础上,井震结合,以测井声波时差和密度曲线建立合成地震记录,以井上合成地震记录为标签,以井旁地震道数据为输入数据,建立井旁地震道数据与井上合成地震记录的训练模型,用来提高地震资料的中高频信息;其次,利用gru实现原始地震记录中低频趋势的保留,将UNet和gru结果组合在一起,实现既提高了中高频信息又保护了地震数据中的低频信息;然后开展模型训练,将模型应用到三维地震数据体上开展计算,提高地震资料分辨率。相对以往方法,提取出的信息更加丰富。通过理论模型和实际数据应用,表明本文方法在提高叠后地震资料分辨率方面有效。

关键词：叠后地震数据测井数据提高分辨率深度学习 UNet gru