Dynamic Economic Scheduling with Self-Adaptive Uncertainty in Distribution Network Based on Deep Reinforcement Learning

Traditional optimal scheduling methods are limited to accurate physical models and parameter settings, which aredifficult to adapt to the uncertainty of source and load, and there are problems such as the inability to make dynamicdecisions continuously. This paper proposed a dynamic economic scheduling method for distribution networksbased on deep reinforcement learning. Firstly, the economic scheduling model of the new energy distributionnetwork is established considering the action characteristics of micro-gas turbines, and the dynamic schedulingmodel based on deep reinforcement learning is constructed for the new energy distribution network system with ahigh proportion of new energy, and the Markov decision process of the model is defined. Secondly, Second, for thechanging characteristics of source-load uncertainty, agents are trained interactively with the distributed networkin a data-driven manner. Then, through the proximal policy optimization algorithm, agents adaptively learn thescheduling strategy and realize the dynamic scheduling decision of the new energy distribution network system.Finally, the feasibility and superiority of the proposed method are verified by an improved IEEE 33-node simulationsystem.

基于向量残差SVD的混凝土超声测试温度效应研究

为研究温度对混凝土超声测试尾波信号的影响规律;将信号向量间的归一化夹角作为波动指标,反映温度效应引起的信号变化;通过向量残差矩阵SVD获得表征温度效应大小的特征向量,建立向量空间映射和温度差的数学关系。在实验室采集混凝土梁尾波信号进行验证,结果表明,随温度升高尾波信号的波形发生后移,文中方法可分段线性量化温度效应;基于量化结果,得到常温下超声尾波信号最敏感的温度区间;任意4.5℃范围内,可去除74%~90%的温度效应。

基于SVD-CWT和CNN的水轮发电机转子故障识别

水轮发电机转子振动故障识别是水电站运维的重难点问题,为此提出一种基于转子振动信号的故障识别方法。首先针对发电机转子的非平稳和非线性振动信号,采用奇异值分解(SVD)并结合能量差分谱理论进行降噪预处理;对预处理数据使用连续小波变换(CWT)转换为时频图并形成图像数据集;然后将该图像数据集作为卷积神经网络(CNN)输入,通过CNN多层池化及卷积形成分布式故障特征表达,最终实现发电机转子故障模式识别和分类。经实验验证,该方法准确率达到99.5%以上,能有效识别出发电机转子的故障类型。

基于自适应动态粒子群优化的RAK-SVD方法

K均值奇异值分解(K-SVD)算法是一种行之有效的地震资料去噪方法,但由于其稀疏分解存在不确定性,需要引入正则项对其改进。为此,在常规粒子群算法的基础上,提出了一种自适应动态粒子群算法优化正则化参数的正则化近似K-SVD(RAK-SVD)去噪方法。首先通过修改字典原子和相关参数,解决了由于常规粒子群算法的惯性参数固定不变,导致后期搜索效率下降的问题;其次将正则化系数引入近似K-SVD(AK-SVD)方法,明显提升了去噪效果;最后利用自适应动态粒子群算法自动优选AK-SVD方法中的正则化参数,提高了稀疏分解的确定性,在对强反射信号进行去噪的同时加强了对弱信号的保护。模型测试和实际应用均表明,该方法有利于弱信号的提取和识别,不仅能够显著改善弱地震信号的去噪效果,还提升了计算效率。该方法具有一定的实际应用价值。

基于CRS-LMD和SVD的MMC-HVDC线路故障测距方法

直流输电线路故障行波波速不确定、波头提取困难以及噪声干扰等因素制约了直流电网中故障测距技术的应用。为了降低上述因素对定位准确性的影响,提出一种基于局部特征有理样条插值均值分解(LMD based on characteristic rational spline,CRS-LMD)和奇异值分解(singular value decomposition,SVD)的故障测距方法。首先,利用特征尺度选取最优极点系数,结合有理样条插值调节拟合曲线的松紧程度,实现对故障电压行波的局部均值分解。其次,采用奇异值分解对故障行波波头进行准确提取。最后,在PSCAD/EMTDC中搭建了张北±500 kV柔性直流电网的仿真模型,模拟各种故障情况并输出故障数据,利用Matlab对故障数据进行处理并验证定位算法。最后,仿真结果表明,所提故障测距算法在不同故障距离和故障类型下均能实现故障测距,且在叠加噪声和过渡电阻的情况下也能保障较高的精确性。

基于SVD-K-means算法的软扩频信号伪码序列盲估计

针对通信中软扩频信号伪码序列盲估计困难的问题,提出一种奇异值分解(singular value decomposition,SVD)和K-means聚类相结合的方法。该方法先对接收信号按照一倍伪码周期进行不重叠分段构造数据矩阵。其次对数据矩阵和相似性矩阵分别进行SVD完成对伪码序列集合规模数的估计、数据降噪、粗分类以及初始聚类中心的选取。最后通过K-means算法优化分类结果,得到伪码序列的估计值。该算法在聚类之前事先确定聚类数目,大大减少了迭代次数。同时实验结果表明,该算法在信息码元分组小于5 bit,信噪比大于-10 dB时可以准确估计出软扩频信号的伪码序列,性能较同类算法有所提升。

Formal Modeling of Self-Adaptive Resource Scheduling in Cloud

A self-adaptive resource provisioning on demand is a critical factor in cloud computing.The selection of accurate amount of resources at run time is not easy due to dynamic nature of requests.Therefore,a self-adaptive strategy of resources is required to deal with dynamic nature of requests based on run time change in workload.In this paper we proposed a Cloud-based Adaptive Resource Scheduling Strategy(CARSS)Framework that formally addresses these issues and is more expressive than traditional approaches.The decision making in CARSS is based on more than one factors.TheMAPE-K based framework determines the state of the resources based on their current utilization.Timed-Arc Petri Net(TAPN)is used to model system formally and behaviour is expressed in TCTL,while TAPAAL model checker verifies the underline properties of the system.

Self-adaptive bulk/surface engineering of Bi_(x)O_(y)Br_(z) towards enhanced photocatalysis:Current status and future challenges

The bulk/surface states of semiconductor photocatalysts are imperative parameters to maneuver their performance by significantly affecting the key processes of photocatalysis including light absorption,separation of charge carrier,and surface site reaction.Recent years have witnessed the encouraging progress of self-adaptive bulk/surface engineered Bi_(x)O_(y)Br_(z) for photocatalytic applications spanning various fields.However,despite the maturity of current research,the interaction between the bulk/surface state and the performance of Bi_(x)O_(y)Br_(z) has not yet been fully understood and highlighted.In this regard,a timely tutorial overview is quite urgent to summarize the most recent key progress and outline developing obstacles in this exciting area.Herein,the structural characteristics and fundamental principles of Bi_(x)O_(y)Br_(z)for driving photocatalytic reaction as well as related key issues are firstly reviewed.Then,we for the first time summarized different self-adaptive engineering processes over Bi_(x)O_(y)Br_(z)followed by a classification of the generation approaches towards diverse Bi_(x)O_(y)Br_(z)materials.The features of different strategies,the up-to-date characterization techniques to detect bulk/surface states,and the effect of bulk/surface states on improving the photoactivity of Bi_(x)O_(y)Br_(z)in expanded applications are further discussed.Finally,the present research status,challenges,and future research opportunities of self-adaptive bulk/surface engineered Bi_(x)O_(y)Br_(z)are prospected.It is anticipated that this critical review can trigger deeper investigations and attract upcoming innovative ideas on the rational design of Bi_(x)O_(y)Br_(z)-based photocatalysts.

A WSN Node Fault Diagnosis Model Based on BRB with Self-Adaptive Quality Factor

Wireless sensor networks (WSNs) operate in complex and harshenvironments;thus, node faults are inevitable. Therefore, fault diagnosis ofthe WSNs node is essential. Affected by the harsh working environment ofWSNs and wireless data transmission, the data collected by WSNs containnoisy data, leading to unreliable data among the data features extracted duringfault diagnosis. To reduce the influence of unreliable data features on faultdiagnosis accuracy, this paper proposes a belief rule base (BRB) with a selfadaptivequality factor (BRB-SAQF) fault diagnosis model. First, the datafeatures required for WSN node fault diagnosis are extracted. Second, thequality factors of input attributes are introduced and calculated. Third, themodel inference process with an attribute quality factor is designed. Fourth,the projection covariance matrix adaptation evolution strategy (P-CMA-ES)algorithm is used to optimize the model’s initial parameters. Finally, the effectivenessof the proposed model is verified by comparing the commonly usedfault diagnosis methods for WSN nodes with the BRB method consideringstatic attribute reliability (BRB-Sr). The experimental results show that BRBSAQFcan reduce the influence of unreliable data features. The self-adaptivequality factor calculation method is more reasonable and accurate than thestatic attribute reliability method.

基于改进SVD++算法和K-means++算法的小文件合并方案

文章提出了一种基于改进SVD++算法和K-means++算法的小文件合并方案。通过引入自适应学习率函数和基于并行分组的SVD++算法,优化了小文件的合并过程,以提高Hadoop存储小文件的效率。同时,利用K-means++算法对合并后的文件进行聚类,优化了数据存储方式,降低了存储空间的浪费。在Hadoop平台上进行的实验表明,该方案在保持数据处理准确性和稳定性的同时,显著提升了Hadoop存储与处理小文件的性能。

基于K-SVD算法的数字图像自适应修复方法

为了提升数字图像的完整性和清晰度,提出一种基于K-SVD算法的数字图像自适应修复方法。通过FCM算法将数字图像划分成不同的图像块,将不同类别的数字图像依据K-SVD算法的稀疏编码和字典更新模块进行训练,获取各个不同类别数字图像块的字典,求出其稀疏系数,结合字典和稀疏系数更新数字图像中的每一类图像块,完成数字图像中每一类图像块的修复或重构,将修复好的图像块放回原数字图像中,实现数字图像的自适应修复。实验结果表明,该方法能够有效地恢复图像的细节和结构,修复后的数字图像均方根误差低,并且具有较高的峰值信噪比,同时,修复后的数字图像与原图像的结构相似性高达0.95,且在数字图像修复效率方面具备显著优势。

利用LMD-SVD方法进行GNSS坐标时间序列降噪

为降低噪声对GNSS坐标时间序列的影响、有效提取时间序列中的有用信息,在局部均值分解(LMD)降噪方法的基础上引入奇异值分解(SVD)方法,建立了LMD-SVD方法。首先通过LMD方法将时间序列分解为若干个乘积函数(PF)和余量,PF分量可反映时间序列的时频分布特性;然后通过连续均方根误差方法确定高频分量与低频分量的分界点;最后对经SVD方法降噪后的高频分量、低频分量和余量进行重构,得到最终降噪结果。利用5个GNSS测站U方向坐标时间序列对该方法进行验证。结果表明,相较于单一LMD方法,LMD-SVD方法结果的信噪比与相关系数分别提高了34.28%与17.11%,均方根误差降低了51.31%,降噪效果更好。

二级减速器故障系统建模及SVD-MMSE劣化评估

为检测故障齿轮劣化程度并进行有效的程度评估,通过有限元法建立含有正常、裂纹和断齿等3种齿轮状态的二级直齿轮减速器系统模型。首先,分别计算3种状态的齿轮时变啮合刚度,并综合考虑轴承支撑刚度,得到了3种不同状态下的轴承振动响应;其次,引入多元多尺度样本熵(multivariate multiscale sample entropy,简称MMSE)对故障齿轮的劣化程度进行分析;最后,引进奇异值分解(singular value decomposition,简称SVD)算法进行预处理,以达到更好的诊断效果来综合评定故障齿轮生命周期的劣化程度。结果表明:齿轮发生故障时,主要导致时频域信号发生转频调制,时域存在有规律的冲击,频域出现边频带,且分布在输入轴的转频及其倍频和啮频及其倍频处;随着故障程度的增加,劣化越发明显,频率成分也发生改变,致使MMSE值也随之变化,且整体呈单调递减趋势;SVD-MMSE算法能有效地对齿轮故障程度进行判别,降低了噪声对于劣化程度检测准确性的影响。

基于SVD配准的空间圆拟合方法

针对传统空间圆拟合方法精度低、方法不稳定等问题,提出一种基于SVD配准解算旋转、平移及缩放系数新方法,进而解算空间圆半径、圆心坐标、圆度和平整度等空间形态参数。与采用最小二乘方法拟合空间圆进行对比,该方法无须迭代,仅需一个缩放系数初始值即可解算空间形态参数,更为快捷高效。为保证计算精度,利用均值漂移的后验概率实现粗差剔除,并以某地铁隧道钢环测量工程实例验证该算法的可靠性与可行性。

In-situ interfacial passivation and self-adaptability synergistically stabilizing all-solid-state lithium metal batteries

The function of solid electrolytes and the composition of solid electrolyte interphase(SEI)are highly significant for inhibiting the growth of Li dendrites.Herein,we report an in-situ interfacial passivation combined with self-adaptability strategy to reinforce Li_(0.33)La_(0.557)TiO_(3)(LLTO)-based solid-state batteries.Specifically,a functional SEI enriched with LiF/Li_(3)PO_(4) is formed by in-situ electrochemical conversion,which is greatly beneficial to improving interface compatibility and enhancing ion transport.While the polarized dielectric BaTiO_(3)-polyamic acid(BTO-PAA,BP)film greatly improves the Li-ion transport kinetics and homogenizes the Li deposition.As expected,the resulting electrolyte offers considerable ionic conductivity at room temperature(4.3 x 10~(-4)S cm^(-1))and appreciable electrochemical decomposition voltage(5.23 V)after electrochemical passivation.For Li-LiFePO_(4) batteries,it shows a high specific capacity of 153 mA h g^(-1)at 0.2C after 100 cycles and a long-term durability of 115 mA h g^(-1)at 1.0 C after 800 cycles.Additionally,a stable Li plating/stripping can be achieved for more than 900 h at 0.5 mA cm^(-2).The stabilization mechanisms are elucidated by ex-situ XRD,ex-situ XPS,and ex-situ FTIR techniques,and the corresponding results reveal that the interfacial passivation combined with polarization effect is an effective strategy for improving the electrochemical performance.The present study provides a deeper insight into the dynamic adjustment of electrode-electrolyte interfacial for solid-state lithium batteries.

基于正则化SVD算法的660MW机组煤粉加热炉炉膛三维温度场重建

针对现有加热炉炉膛内三维温度场重建方法存在的重建误差较大、重建消耗时间较长的问题,提出基于正则化SVD算法的660MW机组煤粉加热炉炉膛三维温度场重建方法。根据人眼视觉二维图像特征点提取原理,提取温度场立体图像特征点;利用小波变换方法计算子线段端点,获取特征点匹配结果;通过声学测温方法以及射线成像理论,重建声波传播速度分布形式,凭借正则化SVD算法构建声学测量系统模型,对声波飞行值进行修正,结合特征点匹配结果和对称轴,得到实现660MW机组煤粉加热炉炉膛三维温度场重建。实验结果表明,所提方法的最低AER、MER、RMSE分别为4.11、0.98、1.21,重建时间始终保持在0.6s以内,重建误差较小、重建消耗时间较短,抗噪声能力强,温度场重建效果好。

基于视觉密码和DCT-SVD彩色图像水印技术

本文提出了一种基于视觉密码和DCT-SVD的彩色图像水印算法,通过视觉密码方案将彩色秘密水印图像拆分成3份与彩色秘密水印图像等大小的分享份图像,并利用待标记图像、水印图像和密钥生成验证信息和测试图像。由版权所有者提供水印图像和验证信息,密钥交由图像所有者保存。当要验证图像所有者的所有权时,会要求图像所有者提供密钥,结合密钥、测试图像和验证信息便可生成水印,若密钥正确,生成的水印可识别为原始水印。将其中一份分享份图像利用DCT-SVD水印算法嵌入到载体图像中并进行攻击测试,利用数字水印的提取算法将分享份图像从遭受常见攻击的载体图像中提取出来,与剩余分享份进行叠加恢复秘密水印图像。实验结果表明,水印具有良好的不可感知性和鲁棒性。

DeepSVDNet:A Deep Learning-Based Approach for Detecting and Classifying Vision-Threatening Diabetic Retinopathy in Retinal Fundus Images

Artificial Intelligence(AI)is being increasingly used for diagnosing Vision-Threatening Diabetic Retinopathy(VTDR),which is a leading cause of visual impairment and blindness worldwide.However,previous automated VTDR detection methods have mainly relied on manual feature extraction and classification,leading to errors.This paper proposes a novel VTDR detection and classification model that combines different models through majority voting.Our proposed methodology involves preprocessing,data augmentation,feature extraction,and classification stages.We use a hybrid convolutional neural network-singular value decomposition(CNN-SVD)model for feature extraction and selection and an improved SVM-RBF with a Decision Tree(DT)and K-Nearest Neighbor(KNN)for classification.We tested our model on the IDRiD dataset and achieved an accuracy of 98.06%,a sensitivity of 83.67%,and a specificity of 100%for DR detection and evaluation tests,respectively.Our proposed approach outperforms baseline techniques and provides a more robust and accurate method for VTDR detection.

一种基于SVD降噪和CNN分类的无线信号调制识别方法

针对低信噪比下噪声干扰导致的调制信号识别精度不足的问题,提出了一种基于奇异值分解(Singular Value Decomposition,SVD)降噪和卷积神经网络(Convolutional Neural Network,CNN)分类的无线信号调制识别方法SVD-CNN。该方法提出了基于SVD的信号降噪模块来对输入信号进行降噪,设计了一维符号级CNN架构来直接识别信号特征并分类。针对高斯、瑞利信道下的调制仿真数据集,将提出的方法与典型调制识别方法如CNN识别方法、瞬时特征-全连接神经网络(Instantaneous Characteristic-Fully Connected Neural Network,IC-FCNN)识别方法进行了对比实验。实验结果表明,所提方法在低信噪比下具有更高的识别精度,在信噪比为0 dB时平均识别准确率提升近38%~49%。

基于SeqGAN和Bi-GRU实施过采样的SVD方法

近年来,基于深度学习的软件漏洞检测方法,在开源C/C++漏洞数据集上取得了97%的准确率。然而,由于漏洞样本占比较低而导致的类不平衡问题,以上方法的误报率和漏报率较高。鉴于SeqGAN的生成策略和Bi-GRU的表征学习能力,提出了基于SeqGAN、Bi-GRU和TextCNN的软件漏洞检测方法。利用SeqGAN对开源C/C++漏洞数据集进行过采样处理以生成过采样漏洞样本,经Bi-GRU挖掘过采样漏洞样本和真实漏洞样本的共有潜在特征,最终提高TextCNN对真实漏洞样本的检测性能。实验结果表明,所提方法在7个开源软件项目的C/C++汇总漏洞数据集上取得了0.9538的F1 score,与TextCNN在应用降采样技术RUS处理后的数据集上的测试结果相比,提高了12.3%;与先进的基于深度学习的软件漏洞检测方法,如VulDeePecker相比,提高了82.5%。