Precipitation Retrieval from Himawari-8 Satellite Infrared Data Based on Dictionary Learning Method and Regular Term Constraint

In this paper,the application of an algorithm for precipitation retrieval based on Himawari-8 (H8) satellite infrared data is studied.Based on GPM precipitation data and H8 Infrared spectrum channel brightness temperature data,corresponding "precipitation field dictionary" and "channel brightness temperature dictionary" are formed.The retrieval of precipitation field based on brightness temperature data is studied through the classification rule of k-nearest neighbor domain (KNN) and regularization constraint.Firstly,the corresponding "dictionary" is constructed according to the training sample database of the matched GPM precipitation data and H8 brightness temperature data.Secondly,according to the fact that precipitation characteristics in small organizations in different storm environments are often repeated,KNN is used to identify the spectral brightness temperature signal of "precipitation" and "non-precipitation" based on "the dictionary".Finally,the precipitation field retrieval is carried out in the precipitation signal "subspace" based on the regular term constraint method.In the process of retrieval,the contribution rate of brightness temperature retrieval of different channels was determined by Bayesian model averaging (BMA) model.The preliminary experimental results based on the "quantitative" evaluation indexes show that the precipitation of H8 retrieval has a good correlation with the GPM truth value,with a small error and similar structure.

Exponential Attractors of the Nonclassical Diffusion Equations with Lower Regular Forcing Term

In this article, we prove the existence of exponential attractors of the nonclassical diffusion equation with critical nonlinearity and lower regular forcing term. As an additional product, we show that the fractal dimension of the global attractors of this problem is finite.

ON A REGULARIZATION OF INDEX 2 DIFFERENTIAL-ALGEBRAIC EQUATIONS WITH PROPERLY STATED LEADING TERM

In this article, linear regular index 2 DAEs A（t）[D（t）x（t）]＇ ＋ B（t）x（t） = q（t） are considered. Using a decoupling technique, initial condition and boundary condition are properly formulated. Regular index 1 DAEs are obtained by a regularization method. We study the behavior of the solution of the regularization system via asymptotic expansions. The error analysis between the solutions of the DAEs and its regularization system is given.

TWO REGULARIZATION METHODS FOR IDENTIFYING THE SOURCE TERM PROBLEM ON THE TIME-FRACTIONAL DIFFUSION EQUATION WITH A HYPER-BESSEL OPERATOR

In this paper,we consider the inverse problem for identifying the source term of the time-fractional equation with a hyper-Bessel operator.First,we prove that this inverse problem is ill-posed,and give the conditional stability.Then,we give the optimal error bound for this inverse problem.Next,we use the fractional Tikhonov regularization method and the fractional Landweber iterative regularization method to restore the stability of the ill-posed problem,and give corresponding error estimates under different regularization parameter selection rules.Finally,we verify the effectiveness of the method through numerical examples.

An improved four-dimensional variation source term inversion model with observation error regularization

Aiming at the Four-Dimensional Variation source term inversion algorithm proposed earlier,the observation error regularization factor is introduced to improve the prediction accuracy of the diffusion model,and an improved Four-Dimensional Variation source term inversion algorithm with observation error regularization(OER-4DVAR STI model)is formed.Firstly,by constructing the inversion process and basic model of OER-4DVAR STI model,its basic principle and logical structure are studied.Secondly,the observation error regularization factor estimation method based on Bayesian optimization is proposed,and the error factor is separated and optimized by two parameters:error statistical time and deviation degree.Finally,the scientific,feasible and advanced nature of the OER-4DVAR STI model are verified by numerical simulation and tracer test data.The experimental results show that OER-4DVAR STI model can better reverse calculate the hazard source term information under the conditions of high atmospheric stability and flat underlying surface.Compared with the previous inversion algorithm,the source intensity estimation accuracy of OER-4DVAR STI model is improved by about 46.97%,and the source location estimation accuracy is improved by about 26.72%.

Impact Force Localization and Reconstruction via ADMM-based Sparse Regularization Method

In practice,simultaneous impact localization and time history reconstruction can hardly be achieved,due to the illposed and under-determined problems induced by the constrained and harsh measuring conditions.Although l_(1) regularization can be used to obtain sparse solutions,it tends to underestimate solution amplitudes as a biased estimator.To address this issue,a novel impact force identification method with l_(p) regularization is proposed in this paper,using the alternating direction method of multipliers(ADMM).By decomposing the complex primal problem into sub-problems solvable in parallel via proximal operators,ADMM can address the challenge effectively.To mitigate the sensitivity to regularization parameters,an adaptive regularization parameter is derived based on the K-sparsity strategy.Then,an ADMM-based sparse regularization method is developed,which is capable of handling l_(p) regularization with arbitrary p values using adaptively-updated parameters.The effectiveness and performance of the proposed method are validated on an aircraft skin-like composite structure.Additionally,an investigation into the optimal p value for achieving high-accuracy solutions via l_(p) regularization is conducted.It turns out that l_(0.6)regularization consistently yields sparser and more accurate solutions for impact force identification compared to the classic l_(1) regularization method.The impact force identification method proposed in this paper can simultaneously reconstruct impact time history with high accuracy and accurately localize the impact using an under-determined sensor configuration.

基于极限学习机的短期电力负荷在线预测

为满足智能电网对电力负荷实时预测的需求,提出基于稀疏递归最小二乘法的极限学习机(SRLS-ELM)在线学习算法,用于短期电力负荷的在线预测.相比在线序列ELM, SRLS-ELM算法无需选择离线样本初始化网络输出权重,将网络学习的平方误差与输出权值的稀疏正则化项相结合,用l1-范数稀疏化网络隐藏层节点,用次梯度策略解决求解过程中代价函数无法处处可微的问题,以递归最小二乘的训练方法完成在线学习,根据估计误差自适应寻找最优正则化参数.仿真结果表明,基于SRLS-ELM的算法能有效简化网络结构,且与ELM、堆叠核ELM批量、在线序列ELM半在线以及精确在线支持向量机回归模型相比,对短期电力负荷在线预测时具有更高的预测精度和学习效率,且鲁棒性强.

基于MRSDAE-KPCA结合Bi-LST的滚动轴承剩余使用寿命预测

针对现有滚动轴承剩余使用寿命预测方法在提取数据特征时没有充分考虑数据的内部分布,且在构建健康因子时还需要专家经验进行人工提取等问题,提出一种基于流形正则化堆栈去噪自编码器、核主成分分析并结合双向长短时记忆网络的滚动轴承剩余使用寿命预测方法。首先采用无监督的堆栈去噪自编码器网络对原始振动数据进行深层特征提取,并使用核主成分分析法进一步降维,以提高健康因子的指标稳定性;然后在堆栈去噪自编码器中加入流形正则化,最大程度保留编码器隐藏层内部的数据分布结构,提高模型提取数据特征的有效性。最后使用双向长短时记忆网络预测轴承的剩余使用寿命,并采用AdaMax优化算法对网络模型的超参数进行自适应寻优。分析结果表明,提出的滚动轴承剩余使用寿命预测方法具有更高的精度。

两阶段非负矩阵分解算法及其在光谱解混中的应用

非负矩阵分解问题(nonnegative matrix factorization,NMF)模型已成功应用至高光谱遥感影像处理中的光谱解混工作,由于NMF优化模型具有多个局部极小点,使得分解结果不稳定。设计初始化方法或者选择带正则项的问题模型是提高分解精度的两种常用方法。本文提出了两阶段的NMF算法,实现了初始点选取和正则项设计的结合。第一阶段借助k-均值获得k个聚类中心,给出迭代的初始点;利用第一阶段的初始矩阵U^(0),定义了针对端元矩阵的正则项‖U-U^(0)‖_(F)^(2),第二阶段采用基于交替非负最小二乘框架的投影梯度算法,求解新的正则化NMF问题。正则项中的端元初始矩阵U^(0)除了采用k-均值获得k个聚类中心,也可采用真实地物光谱,它的引入提高了算法的灵活度。数值结果表明新算法更加稳定,且分解的精确性有效提高。

自适应图嵌入和非凸正则特征自表达的无监督特征选择

针对传统的无监督特征选择不能充分兼顾样本及特征的局部结构,以及没有考虑非凸正则项带来更稀疏的解并能够选择出更具判别性特征等问题,提出了自适应图嵌入和非凸正则特征自表达的无监督特征选择方法。通过图嵌入降低特征维度,获得样本相似度矩阵,引导特征选择;引入特征自表达策略,用其余特征线性表示每一个特征,考虑特征间的相似性关系,保持特征局部结构;在特征自表达中添加非凸正则项,获得行更稀疏的权重矩阵,实现特征选择;在特征选择过程中执行自适应图嵌入对数据局部结构进行学习,选择最优特征子集;为求解非凸稀疏问题,使用交替迭代方法优化求解模型,设计了一种新的无监督特征选择算法。在6个数据集上与其他算法进行实验对比分析,实验结果表明所提算法是有效的。

基于ST-UNet和目标特征的混凝土裂缝检测

混凝土裂缝对结构承载能力、耐久性和防水性有负面影响,因此早期裂缝检测具有重要意义.大数据和深度学习的快速发展,为裂缝智能检测提供了有效的方法.针对裂缝检测过程中图像正负样本不平衡,裂缝区域色彩深沉和低亮度的特点,提出一种基于ST-UNet(Swin Transformer U-Net)和目标特征的裂缝检测方法.该算法在网络中引入CBAM注意力机制,使网络更加关注图像中对裂缝检测起决定作用的像素区域,增强裂缝图像的特征表达能力;使用Focal+Dice混合损失函数代替单一交叉熵损失函数,处理样本图像正负样本分布不均的问题;设计APSD正则化项优化损失函数,针对裂缝区域色彩深沉、低亮度的问题,降低检测的漏检率与误检率.裂缝检测结果表明:IoU指标提升22%,Dice指数提升17%,该算法是有效可行的.

基于Edge-TB的联邦学习中客户端选择策略和数据集划分研究

联邦学习是分布式机器学习在现实中的应用之一。针对联邦学习中的异构性,基于FedProx算法,提出优先选择近端项较大的客户端选择策略,效果优于常见的选择局部损失值较大的客户端选择策略,可以有效提高FedProx算法在异构数据和系统下的收敛速度,提高有限聚合次数内的准确率。针对联邦学习数据异构的假设,设计了一套异构数据划分流程,得到了基于真实图像数据集的异构联邦数据集作为实验数据集。使用开源的分布式机器学习框架Edge-TB作为实验测试平台,以异构划分后的Cifar10作为数据集,实验表明,采用新的客户端选择策略的改进FedProx算法较原算法在有限的聚合轮数内准确率提升14.96%,通信开销减小6.3%;与SCAFFOLD算法相比,准确率提升3.6%,通信开销减小51.7%,训练时间减少15.4%。

用于人脸情感识别的低冗余稀疏性集成剪枝

为降低噪声和异常值对集成剪枝的影响,并鲁棒性地选择更稀疏的基分类器子集,从而提升人脸情感识别的性能,提出了一种具有依赖分数的鲁棒性稀疏低冗余集成剪枝方法用于人脸情感识别。首先,该方法将样本实例的预测结果视为基分类器特征,分别利用互信息和熵来评估成对基分类器之间的依赖性和它们之间的优先级。其次,将优先级依赖关系添加到基于回归的目标方程中实现冗余基分类器的修剪,此目标方程使用l_(2,1)范数来增加分类器子集的鲁棒性从而提升算法的泛化性能。然后,将内积正则化项引入到目标方程中,通过计算分类器特征系数向量内积的绝对值的和去选择稀疏和低冗余的基分类器。最后,使用大多数投票法对选择的基分类器子集进行集成从而得到最终的识别结果。结果表明:本文提出的方法在FER2013、JAFFE、CK+和KDEF 4个公共人脸情感数据集上的识别准确率,比所有基分类器进行集成得到的准确率分别高3.29%、10.39%、1.76%和4.89%,表明该方法可以选择出识别效果更好、冗余度更低的分类器子集,提高集成剪枝的泛化能力。

长期服役锅炉检验案例分析

长期服役的锅炉各部件材料会逐步出现老化现象,特别是在蒸汽参数较高的部位。对于这种锅炉的定期检验,采用当下实施的检验检测标准进行,同时还要考虑锅炉设计和施工的标准。因此在检验中,必须使用各种检验方法,确保检验结果准确无误。本文以主蒸汽管道作为研究对象,分析检验过程中所遇问题与相应解决方案。

基于深度学习的无线频谱感知算法优化

文章提出一种基于深度学习的无线频谱感知方法,并引入L1正则化方法进行优化。首先,分析无线频谱感知的基本原理;其次,设计基于循环神经网络(Recurrent Neural Network,RNN)的频谱感知模型,并对传统RNN模型进行L1正则化优化;最后,利用单输入多输出(Single Input Multiple Output,SIMO)协同频谱感知场景数据集进行实验验证。实验结果表明,相较于传统RNN方法,优化后的RNN模型在准确率、召回率以及F1值等指标上均实现了更好的性能表现,证明了文章提出的方法在提高频谱感知性能方面具有显著的优势。

视觉光流矢量场估计算法综述

主要介绍了光流的理论框架及应用背景,描述了主流的光流算法及逐步演变的数学求解模型.重点从保边平滑、抗光照变化影响、大位移光流、异质点滤除、实时性计算等5个不同的角度全面分析了光流模型优化求解中的技术性重点及难点问题,详细介绍了已有的解决策略并对今后的研究方向进行了展望,指出大位移光流、抗光照变化影响、实时性计算等方向将成为将来的研究热点.

非标准热传导方程的热源识别反问题

探论一类含对流项的热传导方程的只含有空间变量的热源识别反问题.这类问题是不适定的,即问题的解不连续依赖于测量数据.利用简化的Tikhonov正则化方法,得到问题的一个正则近似解,并且给出正则解和精确解之间具有Hlder型的误差估计.

两项式弹性波阻抗反演方法在深层碳酸盐岩储层预测中的应用

川东北地区长兴组碳酸盐岩储层埋藏深度大,获取大角度地震信息困难,地震资料信噪比较低。为此,本文通过引入新的近似式,从中、小角度两个弹性阻抗数据体中提取岩石物性参数。文中首先以佐普里兹方程的Aki-Richards近似式为基础,详细推导了以纵横波阻抗表示的新的两项型弹性波阻抗公式,该式不仅考虑了密度与其他地球物理参数的联系,而且在舍弃密度项时,利用纵波速度与密度幂指数拟合关系式使密度近似为纵波阻抗更加合理;其次引入Tikhonov正则化方法以提高该方法反演的精度、稳定性和抗噪性。数值试算与实际应用结果都证明了该方法的可行性。

基于离散Fréchet距离和LS-SVM的短期负荷预测

针对现有电力系统短期负荷预测精度低、数据处理量大、易受输入变量的影响等问题,提出了一种将离散Fréchet距离与LS-SVM相结合的短期负荷预测方法。分析总结了East-Slovakia Power Distribution Company提供的历年负荷数据,结合该地区的用电规律,通过引入离散Fréchet距离,建立离散曲线相似性的数学模型,选取出与基准曲线形状相似的相似日,利用相似日负荷数据对LS-SVM预测模型进行训练。经过仿真验证,并与标准LS-SVM模型得到的结果对比,所提预测方法明显提高了预测精度。

结合局部能量与改进的符号距离正则项的图像目标分割算法

针对传统C-V模型对颜色不均匀图像分割失败并且对初始轮廓和位置敏感问题,以及现有符号距离正则项存在周期性振荡和局部极值问题。该文提出结合局部能量信息和改进的符号距离正则项的图像目标分割算法。首先,将全局图像信息扩展到HSV空间,并使用局部能量项信息分析每个像素及其领域内的统计特性,从而在较少的迭代次数内有效分割颜色分布不均匀图像。其次,改进现有符号距离正则项,改进后的符号距离正则项在避免水平集函数的重新初始化的同时,提高了计算效率,保证了水平集函数演化过程的稳定性。然后,定义阈值判断法的水平集函数演化的终止准则,使曲线准确演化到目标轮廓。该算法与同类模型的对比实验表明该模型具有较高的分割精度和对初始轮廓的鲁棒性。