采用CNN和Bidirectional GRU的时间序列分类研究

时间序列数据具有非离散性、数据之间的时序相关性、特征空间维度大等特点,当前大多数分类方法需要经过复杂的数据处理或特征工程,未考虑到时间序列具有不同时间尺度特征以及序列数据之间的时序依赖。通过结合卷积神经网络和循环神经网络中的双向门控循环单元,提出了一个新的端对端深度学习神经网络模型BiGRU-FCN,不需要对数据进行复杂的预处理,并且通过不同的网络运算来获取多种特征信息,如卷积神经网络在时序信息上的空间特征以及双向循环神经网络在序列上的双向时序依赖特征,对单维时间序列进行分类。在大量的基准数据集上对模型进行实验与评估,实验结果表明,与现有的多种方法相比,所提出的模型具有更高的准确率,具有很好的分类效果。

Applying Stack Bidirectional LSTM Model to Intrusion Detection

Nowadays,Internet has become an indispensable part of daily life and is used in many fields.Due to the large amount of Internet traffic,computers are subject to various security threats,which may cause serious economic losses and even endanger national security.It is hoped that an effective security method can systematically classify intrusion data in order to avoid leakage of important data or misuse of data.As machine learning technology matures,deep learning is widely used in various industries.Combining deep learning with network security and intrusion detection is the current trend.In this paper,the problem of data classification in intrusion detection system is studied.We propose an intrusion detection model based on stack bidirectional long short-term memory(LSTM),introduce stack bidirectional LSTM into the field of intrusion detection and apply it to the intrusion detection.In order to determine the appropriate parameters and structure of stack bidirectional LSTM network,we have carried out experiments on various network structures and parameters and analyzed the experimental results.The classic KDD Cup’1999 dataset was selected for experiments so that we can obtain convincing and comparable results.Experimental results derived from the KDD Cup’1999 dataset show that the network with three hidden layers containing 80 LSTM cells is superior to other algorithms in computational cost and detection performance due to stack bidirectional LSTM model’s ability to review time and correlate with connected records continuously.The experiment shows the effectiveness of stack bidirectional LSTM network in intrusion detection.

Bayonet-corpus:a trajectory prediction method based on bayonet context and bidirectional GRU

Predicting travel trajectory of vehicles can not only provide personalized services to users,but also have a certain effect on traffic guidance and traffic control.In this paper,we build a Bayonet-Corpus based on the context of traffic intersections,and use it to model a traffic network.Besides,Bidirectional Gated Recurrent Unit(Bi-GRU)is used to predict the sequence of traffic intersections in one single trajectory.Firstly,considering that real traffic networks are usually complex and disorder and cannot reflect the higher dimensional relationship among traffic intersections,this paper proposes a new traffic network modeling algorithm based on the context of traffic intersections:inspired by the probabilistic language model,a Bayonet-Corpus is constructed from traffic intersections in real trajectory sequence,so the high-dimensional similarity between corpus nodes can be used to measure the semantic relation of real traffic intersections.This algorithm maps vehicle trajectory nodes into a high-dimensional space vector,blocking complex structure of real traffic network and reconstructing the traffic network space.Then,the bayonets sequence in real traffic network is mapped into a matrix.Considering the trajectories sequence is bidirectional,and Bi-GRU can handle information from forward and backward simultaneously,we use Bi-GRU to bidirectionally model the trajectory matrix for the purpose of prediction.

Multivariety and multimanufacturer drug identification based on near-infrared spectroscopy and recurrent neural network

Near-infrared(NIR)spectral analysis,which has the advantages of rapidness,nondestruction and high-efficiency,is widely used in the detection of feed,food and mineral.In terms of qualitative identification,it can also be used for the discriminant analysis of medicines.Long short-term memory(LSTM)neural network,bidirectional long short-term memory(BiLSTM)neural network and gated recurrent unit(GRU)network are variants of the recurrent neural network(RNN).The potential relationship between nonlinear features learned from the sequence by these variants is used to complete the missions infields such as natural language processing,signal classification and video analysis.Since the effect of these variants in drug identification is still to be studied,this paper constructs a multiclassifier of these three variants,using compoundα-keto acid tablets produced by four manufacturers and repaglinide tablets produced by five manufacturers as the research object.Then,the paper analyzes the impacts of seven different preprocessed methods on the drug NIR data by constructing different layers of LSTM,BiLSTM and GRU networks and compares different classification model indicators and training time of each model.When the spectrum data are pre-processed by z-score normalization,the GRU-3 model has the best accuracy in all models.The BiLSTM models are better for analyzing high coincidence data.The method proposed in this paper can be further extended to other NIR spectroscopy data sets.

STABC-IR:An air target intention recognition method based on bidirectional gated recurrent unit and conditional random field with space-time attention mechanism

The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention Recognition(IR)method for air targets has shortcomings in temporality,interpretability and back-and-forth dependency of intentions.To address these problems,this paper designs a novel air target intention recognition method named STABC-IR,which is based on Bidirectional Gated Recurrent Unit(Bi GRU)and Conditional Random Field(CRF)with Space-Time Attention mechanism(STA).First,the problem of intention recognition of air targets is described and analyzed in detail.Then,a temporal network based on Bi GRU is constructed to achieve the temporal requirement.Subsequently,STA is proposed to focus on the key parts of the features and timing information to meet certain interpretability requirements while strengthening the timing requirements.Finally,an intention transformation network based on CRF is proposed to solve the back-and-forth dependency and transformation problem by jointly modeling the tactical intention of the target at each moment.The experimental results show that the recognition accuracy of the jointly trained STABC-IR model can reach 95.7%,which is higher than other latest intention recognition methods.STABC-IR solves the problem of intention transformation for the first time and considers both temporality and interpretability,which is important for improving the tactical intention recognition capability and has reference value for the construction of command and control auxiliary decision-making system.

Seismic-inversion method for nonlinear mapping multilevel well–seismic matching based on bidirectional long short-term memory networks

In this paper,the recurrent neural network structure of a bidirectional long shortterm memory network(Bi-LSTM)with special memory cells that store information is used to characterize the deep features of the variation pattern between logging and seismic data.A mapping relationship model between high-frequency logging data and low-frequency seismic data is established via nonlinear mapping.The seismic waveform is infinitely approximated using the logging curve in the low-frequency band to obtain a nonlinear mapping model of this scale,which then stepwise approach the logging curve in the high-frequency band.Finally,a seismic-inversion method of nonlinear mapping multilevel well–seismic matching based on the Bi-LSTM network is developed.The characteristic of this method is that by applying the multilevel well–seismic matching process,the seismic data are stepwise matched to the scale range that is consistent with the logging curve.Further,the matching operator at each level can be stably obtained to effectively overcome the problems that occur in the well–seismic matching process,such as the inconsistency in the scale of two types of data,accuracy in extracting the seismic wavelet of the well-side seismic traces,and multiplicity of solutions.Model test and practical application demonstrate that this method improves the vertical resolution of inversion results,and at the same time,the boundary and the lateral characteristics of the sand body are well maintained to improve the accuracy of thin-layer sand body prediction and achieve an improved practical application effect.

融合CNN-BiGRU和注意力机制的网络入侵检测模型

为提高网络入侵检测模型特征提取能力和分类准确率,提出了一种融合双向门控循环单元(CNN-BiGRU)和注意力机制的网络入侵检测模型.使用CNN有效提取流量数据集中的非线性特征;双向门控循环单元(BiGRU)提取数据集中的时序特征,最后融合注意力机制对不同类型流量数据通过加权的方式进行重要程度的区分,从而整体提高该模型特征提取与分类的性能.实验结果表明:其整体精确率比双向长短期记忆网络(BiLSTM)模型提升了2.25%.K折交叉验证结果表明:该模型泛化性能良好,避免了过拟合现象的发生,印证了该模型的有效性与合理性.

基于KPCA-CNN-DBiGRU模型的短期负荷预测方法

本文针对已有神经网络模型在短期负荷预测中输入维度过高、预测误差较大等问题,提出了一种结合核主成分分析、卷积神经网络和深度双向门控循环单元的短期负荷预测方法。先运用核主成分分析法对原始高维输入变量进行降维,再通过卷积深度双向门控循环单元网络模型进行负荷预测。以第九届全国电工数学建模竞赛试题A题中的负荷数据作为实际算例,结果表明所提方法较降维之前预测误差大大降低,与已有预测方法相比也有大幅的误差降低。

基于特征选择及ISSA-CNN-BiGRU的短期风功率预测

针对风电功率随机性大、平稳性低,以及直接输入预测模型往往难以取得较高精度等问题,提出了一种基于特征选择及改进麻雀搜索算法(ISSA)优化卷积神经网络-双向门控循环单元(CNN-BiGRU)的短期风电功率预测方法。首先,利用变分模态分解(VMD)将原始功率分解为一组包含不同信息的子分量,以降低原始功率序列的非平稳性,提升可预测性,同时通过观察中心频率方式确定模态分解数。其次,对每一分量采用随机森林(RF)特征重要度的方法进行特征选择,从风速、风向、温度、空气密度等气象特征因素中,选取对各个分量预测贡献度较高的影响因素组成输入特征向量。然后,建立各分量的CNN-BiGRU预测模型,针对神经网络算法参数难调、手动配置参数随机性大的问题,利用ISSA对模型超参数寻优,自适应搜寻最优参数组合。最后,叠加各分量的预测值,得到最终的预测结果。以中国内蒙古某风电场实际数据进行仿真实验,与多种单一及组合预测方法进行对比,结果表明,本文所提方法相比于其他方法具有更高的预测精度,其平均绝对百分比误差值达到2.644 0%;在其他4个数据集上进行的模型准确性及泛化性验证结果显示,模型平均绝对百分比误差值分别为4.385 3%、3.174 9%、1.576 1%和1.358 8%,均保持在5.000 0%以内,证明本文所提方法具有较好的预测精度及泛化能力。

基于二次分解双向门控单元新型电力系统超短期负荷预测

在新型电力系统中,电力负荷随机性和波动性较强,现有预测方法难以对其实现高精度预测。为此,提出一种基于二次分解和双向门控循环单元的超短期负荷预测模型。首先,针对电力负荷的强随机性和强波动性,利用自适应噪声完备经验模态分解对电力负荷历史序列进行初步分解,使负荷序列更加平稳。随后,对初步分解得到的强非平稳分量运用连续变分模态分解进行二次分解,降低其预测难度。最后,为充分学习电力负荷的时序特征,在预测过程构建基于双向门控循环单元的超短期电力负荷预测模型。实验结果表明,该模型相较于现有优秀预测模型有更高的预测精度。

基于深度学习的复合电能质量扰动识别方法

精准的电能质量扰动识别是对电能质量扰动事件发生后需要解决的主要问题之一,这对划分责任和加快电力市场化进程均具有重要意义,而海量的电能质量监测数据则为电能质量扰动识别提供了条件与机遇。不同的电能质量扰动类型,其电气特征上也存在区别,故可利用不同电能质量扰动波形之间的差异来区分电能质量扰动类型。结合深度学习理论,建立一种基于双向独立循环神经网络的复合电能质量扰动识别方法,通过提取电能质量扰动信号的本质特征量,建立输入序列与输出序列之间的内在对应关系,克服了分析结果对物理特征量的依赖性,提升了电能质量扰动识别准确率。实验结果表明,所提方法可以有效应对复合电能质量扰动的多样性问题,可以直接从原始的底层数据中自主学习复合电能质量扰动信号中所隐藏的本质特征量,识别准确率高。

基于双层注意力和深度自编码器的时间序列异常检测模型

目前时间序列通常具有弱周期性以及高度复杂的相关性特征,传统的时间序列异常检测方法难以检测此类异常。针对这一问题,提出了一种新的无监督时间序列异常检测模型(DA-CBG-AE)。首先,使用新型滑动窗口方法,针对时间序列周期性设置滑动窗口大小;其次,采用卷积神经网络提取时间序列高维度空间特征;然后,提出具有堆叠式Dropout双向门循环单元网络作为自编码器的基本结构,从而捕捉时间序列的相关性特征;最后,引入双层注意力机制,进一步提取特征,选择更加关键的时间序列,从而提高异常检测准确率。为了验证该模型的有效性,将DA-CBG-AE与6种基准模型在8个数据集上进行比较。最终的实验结果表明,DA-CBG-AE获得了最优的F1值(0.863),并且其检测性能相比最新的基准模型Tad-GAN高出25.25%。

基于交叉注意力的多源数据融合的气体泄漏检测

为解决单一传感器在管道气体泄漏检测时容易出现误报、漏报的问题,及时预警并反馈泄漏状况,提出一种基于交叉注意力的多源数据融合管道泄漏检测方法。首先,利用预训练的ShuffleNetV2模型提取热像仪数据的空间特征;然后,结合一维卷积神经网络(1DCNN)和双向门控循环单元(BiGRU),构建1DCNN-BiGRU模型,以提取气体传感器数据的时序特征;最后,运用交叉注意力捕获数据的时空关联性得到2个数据源的特征表示,通过残差方式进行特征连接后输入到分类层中,得到识别结果。结果表明:所构建的多源数据融合模型(SCGA)对气体识别准确率为99.22%,损失值在0~0.04内波动;与仅使用气体传感器数据的支持向量机(SVM)、1DCNN、BiGRU模型相比,准确率至少提升4.12%;与仅使用热图像传感器数据的MobileNetV3、ShuffleNetV2、ResNet18模型相比,准确率至少提升1.14%;与将时序特征和空间特征直接拼接的多源数据融合模型(SCG)相比,准确率提升1%。SCGA模型对气体识别具有较高精度。

基于时序生成对抗网络的居民用户非侵入式负荷分解

现有的非侵入式负荷分解算法往往需要大量电器设备级的负荷数据才能保证分解精度,但由于用户对隐私性的考虑以及安装成本过高等问题,很难获取这些数据。因此,构建一种能深度挖掘电力负荷数据时序特性和电器相关性的时序生成对抗网络。利用降维网络对所有电器有功功率序列组成的高维向量进行降维以降低计算的复杂度,通过复原网络将结果还原为高维向量。基于电器运行状态和深度学习的非侵入式分解方法,运用卷积神经网络-双向门控循环单元构建状态复杂电器的负荷分解回归模型,对状态简单电器利用深度神经网络构建负荷识别分类模型。通过对比其他数据生成方法,以及改变典型公开数据集中生成数据比例所得的负荷分解结果验证了所提方法的有效性。

基于Mogrifier-BiGRU的飞行器轨迹预测

针对当前飞行器轨迹预测准确性低的问题,引入双向传播机制和Mogrifier数据耦合模块,改进传统门控循环单元网络,提出基于Mogrifier-BiGRU的飞行器轨迹预测算法,加强网络对历史数据的学习与记忆,使得输入信息与隐藏层数据的充分耦合,提高预测准确度。仿真结果表明,所提方法对飞行器轨迹预测的准确度可达到96.26%,满足我方作战指挥人员对战场态势趋势准确预测的实际需求。

基于双向循环插补的大地电磁脉冲类噪声处理

【目的】大地电磁测深是一种通过观测天然电磁场获取地下电性结构的勘探方法,较易受到噪声干扰。脉冲类噪声是大地电磁工作中的常见噪声,其幅值高、频带宽,会对数据质量产生较大影响。【方法】为了压制脉冲类噪声,以插补思想为基础,提出了基于时间序列双向循环插补模型(Bidirectional recurrent imputation for time series,BRITS)的大地电磁脉冲类噪声处理方法。首先,将噪声干扰段删除,此时大地电磁时间序列可视为待插补的缺失序列,而后利用该缺失序列构建训练集,对BRITS模型进行插补训练,训练完成后对缺失序列进行插补,即可得到去噪结果。通过仿真及实测含噪声数据处理,并与经验模态分解(Empirical mode decomposition,EMD)阈值方法进行了对比。【结果和结论】结果表明:BRITS方法对仿真噪声数据处理后与原始数据的归一化互相关系数可达0.999以上,信噪比可达29 dB以上,EMD阈值方法处理前后相关系数为0.778,信噪比为3.09 dB;在实测数据处理中,BRITS方法有效恢复了噪声干扰数据,相比EMD阈值方法,其阻抗奈奎斯特图更接近天然大地电磁信号特征。通过不同训练样本试验得出:对4分量大地电磁数据而言,数据中至少需包含两道正常分量,单个含噪分量中噪声占比不大于20%,且噪声连续干扰长度不超过10个采样点,此时,BRITS方法去噪后数据的相关系数在0.96以上,可以保证一定的去噪精度。

改进TCN结合Bi-GRU的人体动作识别方法

针对传统人体动作识别方法特征提取不完善和泛化性能不足导致识别精度不高的问题,提出一种基于深度学习的动作识别模型。改进了传统时域卷积网络(TCN),逐层指数级缩减空洞率,优化了时域卷积的残差结构,实现在浅层网络中提取到长时间间隔数据之间的时域特征和规范网络输出。重构结构进一步结合双向门控循环单元网络(Bi-GRU),提取数据局部特征输入到全连接层整合特征并进行Softmax分类。实验表明,提出的模型在自建数据集和公开数据集UCI-HAR上保持较低参数量的同时,准确率分别达到99.61%和94.16%,具备可靠的识别性能。

基于用户性格和语义-结构特征的文本评论情感分类方法

由于传统文本评论情感分类方法通常忽略用户性格对于情感分类结果的影响,提出一种基于用户性格和语义-结构特征的文本评论情感分类方法(User Personality and Semantic-structural Features based Sentiment Classification Method for Text Comments,BF_Bi GAC).依据大五人格模型能够有效表达用户性格的优势,通过计算不同维度性格得分,从评论文本中获取用户性格特征.利用双向门控循环单元(Bidirectional Gated Recurrent Unit,Bi GRU)和卷积神经网络(Convolutional Neural Network,CNN)可以有效提取文本上下文语义特征和局部结构特征的优势,提出一种基于Bi GRU、CNN和双层注意力机制的文本语义-结构特征获取方法.为区分不同类型特征的影响,引入混合注意力层实现对用户性格特征和文本语义-结构特征的有效融合,以此获得最终的文本向量表达.在IMDB、Yelp-2、Yelp-5及Ekman四个评论数据集上的对比实验结果表明,BF_Bi GAC在分类准确率(Accuracy)和加权macro F_(1)值(F_(w))上均获得较好表现,相对于拼接Bi GRU、CNN的情感分类方法(Sentiment Classification Method Concatenating Bi GRU and CNN,Bi G-RU_CNN)在Accuracy值上分别提升0.020、0.012、0.017及0.011,相对于拼接CNN、Bi GRU的情感分类方法(Sentiment Classification Method Concatenating CNN and Bi GRU,Conv Bi LSTM)F_(w)值上分别提升0.022、0.013、0.028及0.023;相对于预训练模型BERT和Ro BERTa,BF_Bi GAC在保证分类精度的情况下获得了较高的运行效率.

融合1D-CNN与BiGRU的类不平衡流量异常检测

网络流量异常检测是利用各种检测技术分析判断网络流量,发现网络中潜在的攻击,是一种有效的网络安全防护方法。针对高维海量数据和不同攻击类别的网络流量数据不均衡而导致检测准确率低、误报率高的问题,提出一种融合一维卷积神经网络(1D-CNN)和双向门控循环单元(BiGRU)的类不平衡流量异常检测模型。首先,针对类不平衡数据,通过使用改进的合成少数类过采样技术(SMOTE)即Borderline-SMOTE和基于高斯混合模型(GMM)的欠采样聚类技术进行平衡处理;然后,使用1D-CNN提取数据的局部特征,并利用BiGRU更好地提取数据中的时序特征;最后,在UNSW-NB15数据集对所提模型进行验证,所提模型的准确率为98.12%,误报率为1.28%。结果表明,所提模型提高了对少数攻击的识别率,检测精度高于其他经典机器学习和深度学习模型。

基于注意力机制的IWOA-BiGRU超短期风电功率预测

超短期风电功率预测对电力系统调度及大规模风电并网具有重要作用。为得到准确可靠的风电功率预测结果,针对风电功率数据非线性和时序性的特点,提出一种基于IWOA-AT-BiGRU的超短期风电功率预测方法。首先,提出改进鲸鱼优化算法(improved whale optimization algorithm,IWOA)来优化风电功率预测模型的超参数,加速模型收敛,提高预测准确度;然后,在BiGRU中加入注意力机制(AT),AT用来加强重要信息对风功率的影响,BiGRU同时考虑数据的正反向信息,充分挖掘数据的时序特征;最后,通过某风电场实测数据进行实验,结果表明提出的方法预测准确度均高于其他对比模型,具有良好的预测性能。