Conditional Generative Adversarial Network Enabled Localized Stress Recovery of Periodic Composites

Structural damage in heterogeneousmaterials typically originates frommicrostructures where stress concentration occurs.Therefore,evaluating the magnitude and location of localized stress distributions within microstructures under external loading is crucial.Repeating unit cells(RUCs)are commonly used to represent microstructural details and homogenize the effective response of composites.This work develops a machine learning-based micromechanics tool to accurately predict the stress distributions of extracted RUCs.The locally exact homogenization theory efficiently generates the microstructural stresses of RUCs with a wide range of parameters,including volume fraction,fiber/matrix property ratio,fiber shapes,and loading direction.Subsequently,the conditional generative adversarial network(cGAN)is employed and constructed as a surrogate model to establish the statistical correlation between these parameters and the corresponding localized stresses.The stresses predicted by cGAN are validated against the remaining true data not used for training,showing good agreement.This work demonstrates that the cGAN-based micromechanics tool effectively captures the local responses of composite RUCs.It can be used for predicting potential crack initiations starting from microstructures and evaluating the effective behavior of periodic composites.

Conveyor-Belt Detection of Conditional Deep Convolutional Generative Adversarial Network

In underground mining,the belt is a critical component,as its state directly affects the safe and stable operation of the conveyor.Most of the existing non-contact detection methods based on machine vision can only detect a single type of damage and they require pre-processing operations.This tends to cause a large amount of calculation and low detection precision.To solve these problems,in the work described in this paper a belt tear detection method based on a multi-class conditional deep convolutional generative adversarial network(CDCGAN)was designed.In the traditional DCGAN,the image generated by the generator has a certain degree of randomness.Here,a small number of labeled belt images are taken as conditions and added them to the generator and discriminator,so the generator can generate images with the characteristics of belt damage under the aforementioned conditions.Moreover,because the discriminator cannot identify multiple types of damage,the multi-class softmax function is used as the output function of the discriminator to output a vector of class probabilities,and it can accurately classify cracks,scratches,and tears.To avoid the features learned incompletely,skiplayer connection is adopted in the generator and discriminator.This not only can minimize the loss of features,but also improves the convergence speed.Compared with other algorithms,experimental results show that the loss value of the generator and discriminator is the least.Moreover,its convergence speed is faster,and the mean average precision of the proposed algorithm is up to 96.2%,which is at least 6%higher than that of other algorithms.

Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks

This paper proposes a novel approach for generating 3-dimensional complex geological facies models based on deep generative models.It can reproduce a wide range of conceptual geological models while possessing the flexibility necessary to honor constraints such as well data.Compared with existing geostatistics-based modeling methods,our approach produces realistic subsurface facies architecture in 3D using a state-of-the-art deep learning method called generative adversarial networks(GANs).GANs couple a generator with a discriminator,and each uses a deep convolutional neural network.The networks are trained in an adversarial manner until the generator can create "fake" images that the discriminator cannot distinguish from "real" images.We extend the original GAN approach to 3D geological modeling at the reservoir scale.The GANs are trained using a library of 3D facies models.Once the GANs have been trained,they can generate a variety of geologically realistic facies models constrained by well data interpretations.This geomodelling approach using GANs has been tested on models of both complex fluvial depositional systems and carbonate reservoirs that exhibit progradational and aggradational trends.The results demonstrate that this deep learning-driven modeling approach can capture more realistic facies architectures and associations than existing geostatistical modeling methods,which often fail to reproduce heterogeneous nonstationary sedimentary facies with apparent depositional trend.

Adversarial Training-Aided Time-Varying Channel Prediction for TDD/FDD Systems

In this paper, a time-varying channel prediction method based on conditional generative adversarial network(CPcGAN) is proposed for time division duplexing/frequency division duplexing(TDD/FDD) systems. CPc GAN utilizes a discriminator to calculate the divergence between the predicted downlink channel state information(CSI) and the real sample distributions under a conditional constraint that is previous uplink CSI. The generator of CPcGAN learns the function relationship between the conditional constraint and the predicted downlink CSI and reduces the divergence between predicted CSI and real CSI.The capability of CPcGAN fitting data distribution can capture the time-varying and multipath characteristics of the channel well. Considering the propagation characteristics of real channel, we further develop a channel prediction error indicator to determine whether the generator reaches the best state. Simulations show that the CPcGAN can obtain higher prediction accuracy and lower system bit error rate than the existing methods under the same user speeds.

Phase prediction for high-entropy alloys using generative adversarial network and active learning based on small datasets

In this paper,a new machine learning(ML)model combining conditional generative adversarial networks(CGANs)and active learning(AL)is proposed to predict the body-centered cubic(BCC)phase,face-centered cubic(FCC)phase,and BCC+FCC phase of high-entropy alloys(HEAs).Considering the lack of data,CGANs are introduced for data augmentation,and AL can achieve high prediction accuracy under a small sample size owing to its special sample selection strategy.Therefore,we propose an ML framework combining CGAN and AL to predict the phase of HEAs.The arithmetic optimization algorithm(AOA)is introduced to improve the artificial neural network(ANN).AOA can overcome the problem of falling into the locally optimal solution for the ANN and reduce the number of training iterations.The AOA-optimized ANN model trained by the AL sample selection strategy achieved high prediction accuracy on the test set.To improve the performance and interpretability of the model,domain knowledge is incorporated into the feature selection.Additionally,considering that the proposed method can alleviate the problem caused by the shortage of experimental data,it can be applied to predictions based on small datasets in other fields.

High-speed multimode fiber imaging system based on conditional generative adversarial network

The multimode fiber(MMF)has great potential to transmit high-resolution images with less invasive methods in endoscopy due to its large number of spatial modes and small core diameter.However,spatial modes crosstalk will inevitably occur in MMFs,which makes the received images become speckles.A conditional generative adversarial network(GAN)composed of a generator and a discriminator was utilized to reconstruct the received speckles.We conduct an MMF imaging experimental system of transmitting over 1 m MMF with a 50μm core.Compared with the conventional method of U-net,this conditional GAN could reconstruct images with fewer training datasets to achieve the same performance and shows higher feature extraction capability.

基于改进条件生成对抗网络的可控场景生成方法

可再生能源发电具有较强的随机性和波动性,为实现高效可靠的场景建模,提出一种基于改进条件生成对抗网络的可控场景生成方法。提出基于条件生成对抗网络的场景生成框架,结合Transformer的全局注意力机制以及常规卷积和深度可分离卷积的局部泛化机制,设计适用于提取可再生能源发电不同维度特征的网络结构;利用条件生成对抗网络模型建立低维气象特征隐空间和高维可再生能源发电数据之间的映射关系,提出一种可控场景生成方法,并建立随机场景生成、场景约减、极端场景生成和连续日场景生成4种生成策略。基于实际光伏、风电数据和气象数据的仿真结果表明,所提模型与方法能够有效学习可再生能源发电的随机性、时序性、波动性及空间相关性,实现对不同策略下场景的可控生成。

基于条件生成对抗网络的无线传感网络多节点失效修复研究

当前主流的传感节点失效修复主要通过纠删码完成,修复后节点具有更高的空间利用率,但无法有效提升网络寿命。为此,提出基于条件生成对抗网络的无线传感网络多节点数据重构方法,完成失效修复。感知无线传感网络节点,对失效节点展开裁决,确定失效节点位置,并重构节点内数据;将获取的失效节点用于条件生成对抗网络(CGAN)框架中生成器与节点替换网络的训练,通过训练好的生成器,以失效节点为条件,生成未失效节点;为提升修复性能,使用粒子群算法寻优节点替换网络参数,完成节点重构数据置换,实现失效节点的有效修复。结果表明:利用所提方法进行修复时,能耗最高仅为17 J,剩余寿命最低可达到300 h,连通度最高可达到99.2%,具有较好的修复效果。

面向城区的基于图去噪的小区级RSRP估计方法

移动通信系统网络的规划、部署和优化都不同程度依赖于参考信号接收功率(RSRP)估计的准确性。传统上,基站覆盖小区内某信号接收点的RSRP可由对应的无线传播模型估计。在城市环境中,不同小区的无线传播模型需要使用大量RSRP实测数据校正。由于不同小区环境存在差异,经过校正后的模型只适用于对应小区,且小区内的RSRP估计精度低。针对上述问题,将RSRP估计问题转化为图去噪问题,并通过图像处理与深度学习技术得到小区级无线传播模型,不仅能实现小区整体的RSRP估计,且能适用于相似环境小区。首先,通过随机森林回归器逐点预测每个接收点的RSRP,得到整个小区的RSRP估计图;然后,将RSRP估计图和实测RSRP分布图之间的损失视为RSRP噪声图,提出基于条件生成对抗网络(CGAN)的图去噪RSRP估计方法,通过电子环境地图反映小区的环境信息,有效地降低不同小区的RSRP。实验结果表明,在无实测数据的跨小区RSRP预测场景下,所提方法预测RSRP的均方根误差(RMSE)为6.77 dBm,相较于基于卷积神经网络的RSRP估计方法EFsNet下降2.55 dBm;在同小区RSRP预测场景下,相较于EFsNet,模型参数量减小80.3%。

融合迁移学习与CGAN的风电集群功率超短期预测

针对可再生能源不确定性导致电力系统消纳能力不足的问题,提出一种基于条件生成对抗网络与迁移学习融合的风电集群功率超短期预测方法。首先,分析了风电集群功率预测样本模式的不均衡性以及导致的神经网络预测误差偏移现象;其次,构建了条件生成对抗网络修复不均衡问题;最后,采用迁移学习结合时间卷积网络构建了风电集群功率超短期预测模型。测试结果表明,所提方法能够显著提高风电集群功率超短期预测精度。

融合残差SENet的毫米波大规模MIMO信道估计

在户外光线追踪场景下,针对毫米波大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)系统受户外环境噪声干扰导致估计精度低的问题,提出了一种融合残差挤压激励网络(Squeeze-and-Excitation Network,SENet)的条件生成对抗网络的信道估计方法。该方法采用条件生成对抗网络将低分辨率接收信号重建为高分辨率的原始信号完成信道估计,同时在生成器网络中引入SENet网络模块来抑制户外场景下显著性噪声干扰,提高估计精度;最后将残差网络中的残差块添加到SENet的放缩操作后,提高条件生成对抗网络的收敛速度。仿真结果表明,相较于正交匹配追踪算法、卷积神经网络、去噪卷积神经网络和条件生成对抗网络算法,所提方法在户外噪声环境下估计精度平均提高了约2.2 dB,且在高噪声强度下估计精度的提高更为显著。

考虑新能源发电不确定性的含微电网群共享储能优化调度

随着具有不确定性的可再生能源在微电网中渗透率日益提高,微电网新能源消纳所面临的挑战越来越大,而共享储能与微电网群的协同运行是提高其消纳能力的有效手段。为了解决不确定环境下微电网群与共享储能协同调度问题,提出了一种考虑新能源发电不确定性的优化调度模型:首先,利用基于马尔科夫链蒙特卡洛算法的日状态转移过程集合生成方法,和基于条件生成对抗网络的日场景生成方法,生成周净发电功率典型场景。进而使用生成的典型场景作为运行模拟场景,以共享储能运营商为主体,以各微电网为从体,构建一主多从博弈优化调度模型,其中主体调整共享储能容量租赁价格以追求最大利润,从体响应租赁价格,调整租赁容量方案及运行计划以最小化自身用电成本。然后,设计了一种基于启发式算法的分布式迭代求解方法来求解所提模型。最后通过算例验证了所提模型的有效性。

基于条件生成对抗网络的图像翻译综述

图像翻译旨在实现多组不同领域图像间的转换,同时需要约束样本空间与目标空间分布的一致性.文章旨在寻找条件生成对抗网络与图像翻译问题的结合点,首先,介绍了数据集的特点,指出了不同数据集图像翻译难易程度;其次,从数学表达、性质以及目标函数设计方法得出算法实现的不同方式;将现有图像翻译分成3种类别——匹配图像翻译、非匹配图像翻译和多领域图像翻译,并得出不同应用场景所对应的图像翻译类别,即高清任务对应匹配图像翻译,低成本任务对应非匹配图像翻译,多样化任务对应多领域图像翻译;将图像质量评价方法分为主观图像质量评价与客观图像质量评价,并分析客观图像质量评价中全参考图像与无参考图像质量评价的适用范围;最后,总结条件生成对抗网络在图像翻译中的进展,并分析算法后指出了模式崩塌,模型可解释性和少样本等未来所需解决的问题.

基于改进CGAN的海冰SAR-to-Optical影像转换

遥感海冰监测是当前研究热点,通过条件生成对抗网络(CGAN)将海冰SAR影像转换成光学影像,可获得全天时全天候且形象直观的监测数据,但该方法得到的转换结果存在影像模糊、纹理弱化和颜色失真等问题。本文针对以上问题设计了改进的CGAN网络,综合当前的改进方式,新模型在网络结构上加入了空洞空间金字塔模块并设计了加入交叉特征融合模块的跳跃连接,使用结构相似性和L1范数联合损失函数。本文选取东波弗特海地区5景Sentinel-1影像和7景Sentinel-2影像开展实验,实验结果表明,改进CGAN转换的影像具有更好的视觉效果,峰值信噪比(PSNR)提高了3.4 dB,结构相似性(SSIM)提高了0.11,均方根误差(RMSE)降低了13%,并且经过转换后的影像比SAR影像海冰分类结果准确度提高了7.33%。

利用条件生成对抗网络建立曲流河地质模型

【目的】在传统的河道建模方法中,基于目标的方法难以刻画曲流河点坝且条件化困难,多点地质统计学则难以再现河道的连续形态。条件生成对抗网络可以生成满足一定条件的复杂图形,可解决曲流河地质模型建立过程中点坝、河流形态刻画困难及难以条件化的问题。【方法】以鄂尔多斯盆地苏里格气田南部地区某气田为例,开展了基于条件生成对抗网络的曲流河三维建模方法研究。在建模过程中,首先采用Alluvsim建模方法根据工区曲流河特征建立了200个曲流河模型;再通过卷积神经网络对200个模型进行深度学习,提取模型的特征矩阵,利用条件生成对抗网络建立可以生成曲流河模型的生成器;最后以工区井点数据作为输入数据,利用生成器建立满足曲流河复杂形态和井点数据的三维模型。【结果与结论】所建立的模型可以很好地展现曲流河中河道与点坝的三维形态及对应关系。为明确影响模型结果的关键因素,通过对比训练次数与输入数据发现,适当的训练次数(160次)与大量地输入样本(200个)是建立满足工区条件模型的前提。另外,通过对比传统地质建模方法,条件生成对抗网络建模方法可以很好地再现河道沉积体的空间形态,克服传统曲流河建模方法在条件化方面的困难,为曲流河沉积环境的河道砂体建模提供了新的解决思路,建立的曲流河模型可为油田开发阶段提供参考。

基于联邦学习的多源异构网络无数据融合方法

在联合作战体系中,数据作为基础性战略资源发挥着重要的底层支撑作用,数据妥善管理和高效利用是推动作战能力整体跃迁和作战样式深度变革的重要动力。为实现不同作战系统间信息的互联互通,提出一种基于联邦学习的多源异构网络无数据融合方法。从多源数据融合面临的安全性和异构性问题出发,利用条件生成对抗网络提取本地知识和全局分布,集成数据信息;结合局部教师模型-全局模型架构,以无数据知识蒸馏的方式对局部模型知识进行迁移,融合异构网络,细化全局模型,实现不同系统间安全、高质量的信息交互,为智能化指挥信息系统建设提供技术支撑。实验结果表明:该方法在结构化数据和图像数据上具有可行性,整体准确率可达到80%以上。

基于校正条件生成对抗网络的风电场群绿氢储能系统容量配置

以条件生成对抗网络(CGAN)为代表的半监督学习可计及风电波动并生成出力场景集合,生成的数据可输入氢气储能容量配置模型以支撑优化求解。为此,该文首先设计一种校正条件生成对抗网络(CCGAN),并基于风电预测误差构建条件校正器,对预测失准事件和风电爬坡事件下输入生成器的标签信息进行识别和校正;然后,以储能定容的综合成本和各风场弃风成本为目标函数,构建绿氢储能容量配置的多目标优化模型,并引入基于切比雪夫距离的膝区域数学概念,以指导多目标优化算法设计;最后,以新英格兰39节点系统为例进行算例分析,结果表明未经校正的条件信息将导致定容决策偏离实际,而CCGAN能生成计及风电不确定性的高置信出力场景,使得容量配置结果兼顾鲁棒性和经济性。

考虑含储能的三端智能软开关与需求侧响应的主动配电网有功无功协调优化

针对分布式电源(distributed generation,DG)大规模并网导致的配电网电压越限、网络损耗增加和弃风弃光等问题,提出一种考虑含储能(energy storage system,ESS)的三端智能软开关(soft open point,SOP)与需求侧响应(demand response,DR)的主动配电网(active distribution network,ADN)有功无功协调优化策略。首先,计及可再生能源出力的不确定性,基于条件生成对抗网络(conditional generative adversarial networks,CGAN)生成日前场景。然后,对含ESS的三端SOP与DR进行建模,并将其与DG和传统无功补偿设备进行协调优化。构建以网络损耗最小、电压偏差最小、弃风弃光量最小、用户舒适度与用户经济度最大为优化目标的数学模型。进而转化为混合整数二阶锥规划模型,并利用GUROBI进行求解。最后,在改进的IEEE 33和IEEE 69节点配电系统中进行仿真验证。结果表明,计及风光出力的不确定,所提优化策略在兼顾用户舒适度和用户经济度的同时提高了配电网运行的经济性和可靠性。

基于扩张卷积条件生成对抗网络的红外小目标检测

基于深度神经网络的目标检测方法凭借自身强大的建模能力,在通用目标检测任务中取得了良好的表现。然而,在红外小目标信号弱、像素小的本质特征的影响下,深度神经网络层次的加深和池化操作的大量使用导致小目标语义信息丢失,使得现有方法的检测效果并不理想。文中从红外小目标特性这一关键问题出发,提出了一种新颖的基于扩张卷积条件生成对抗网络的目标检测算法。所提方法应用扩张卷积设计了生成网络,充分利用上下文信息建立层与层之间的关联,将红外小目标更多的语义信息保留到深层网络中,增强目标特征,进而提高检测性能。此外,设计了融合通道与空间维度的混合注意力模块,在特征提取时有选择性地放大目标信息,抑制背景信息;设计了自注意关联模块处理层与层之间信息融合过程中产生的语义冲突问题。文中使用多种评价指标将所提网络模型与目前先进的其他红外小目标检测方法进行对比,证明了该方法在复杂背景下目标检测性能的优越性。在公开的SIRST数据集上,所提模型的F分数为64.70%,相比传统方法提高了8.29%,相比深度学习方法提高了7.29%;在公开的ISOS数据集上,所提模型的F分数为64.54%,相比传统方法提高了23.59%,相比深度学习方法提高了6.58%。

基于条件生成对抗插补网络的双重判别器缺失值插补算法

应用中的各种因素可能造成数据缺失,影响后续任务的分析。因此,数据集缺失值的插补尤为重要。相比原本没有插补的处理,错误的插补值也会对分析造成更严重的偏差。针对这种情况,提出新的采用双重判别器的基于条件生成对抗插补网络(C-GAIN)的缺失值插补算法DDC-GAIN(Dual Discriminator based on C-GAIN)。该算法通过一个辅助判别器辅助主判别器判断预测值的真假,即根据一个样本的全局信息判断这个样本生成的真假,更注重特征之间的关系,以此估算预测值。在4个数据集上与5种经典插补算法进行对比实验,结果表明:同样条件下,DDC-GAIN算法在样本量较大时的均方根误差(RMSE)最低;在Default credit card数据集上缺失率为15%时,DDC-GAIN算法的RMSE比次优算法C-GAIN降低了28.99%。这说明利用辅助判别器帮助主判别器学习特征之间的关系是有效的。