Determination of Smoothing Factor for the Inversion of Co-seismic Slip Distribution

For the determination of the smoothing factor (also known as the regularization parameter) in the co-seismic slip distribution inversion, the compromise curve between the model roughness and the data fitting residual is generally used to determine (in order to distinguish the method proposed in this paper, the method is called “L curve” according to its shape). Based on the L-curve, the Eclectic Intersection curve as a new method is proposed to determine the smoothing factor in this paper. The results of the simulated experiment show that the inversion accuracy of the parameters of the seismic slip distribution with the smoothing factor determined by the Eclectic Intersection curve method is better than that of the L curve method. Moreover, the Eclectic Intersection curve method and the L curve method are used to determine the smoothing factor of L’Aquila earthquake and the Taiwan Meinong earthquake slip distribution inversion respectively, and the inversion results are compared and analyzed. The analysis results show that the L’Aquila and the Taiwan Meinong actual earthquake slip distribution results are in the range of other scholars at home and abroad, and compared with the L curve method, the Eclectic Intersection curve method has advantages of high computation efficiency, no need to depend on data fitting degree and more appropriate of smoothing factor and so on.

Reservoir information extraction using a fractional Fourier transform and a smooth pseudo Wigner-Ville distribution

Currently, it is difficult for people to express signal information simultaneously in the time and frequency domains when analyzing acoustic logging signals using a simple-time or frequency-domain method. It is difficult to use a single type of time-frequency analysis method, which affects the feasibility of acoustic logging signal analysis. In order to solve these problems, in this paper, a fractional Fourier transform and smooth pseudo Wigner Ville distribution （SPWD） were combined and used to analyze array acoustic logging signals. The time-frequency distribution of signals with the variation of orders of fractional Fourier transform was obtained, and the characteristics of the time-frequency distribution of different reservoirs under different orders were summarized. Because of the rotational characteristics of the fractional Fourier transform, the rotation speed of the cross terms was faster than those of primary waves, shear waves, Stoneley waves, and pseudo Rayleigh waves. By choosing different orders for different reservoirs according to the actual circumstances, the cross terms were separated from the four kinds of waves. In this manner, we could extract reservoir information by studying the characteristics of partial waves. Actual logging data showed that the method outlined in this paper greatly weakened cross-term interference and enhanced the ability to identify partial wave signals.

A Study on Load Distribution along Contact Lines of Cylindrical Involute Gears

The mathematical model of current methods for determining load distribution along contact lines of the instantaneously engaged teeth of gears has a defect that its numerical solution is not unique. In order to overcome this defect, a new method based on three-dimensional finite element compliance matrix method improved by smoothing method is presented. Calculated examples of helical gears show that the new method can calculate tooth load distribution accurately and agree well with the results of experiments.

Label distribution expression recognition algorithm based on asymptotic truth value

Ambiguous expression is a common phenomenon in facial expression recognition(FER).Because of the existence of ambiguous expression,the effect of FER is severely limited.The reason maybe that the single label of the data cannot effectively describe complex emotional intentions which are vital in FER.Label distribution learning contains more information and is a possible way to solve this problem.To apply label distribution learning on FER,a label distribution expression recognition algorithm based on asymptotic truth value is proposed.Under the premise of not incorporating extraneous quantitative information,the original information of database is fully used to complete the generation and utilization of label distribution.Firstly,in training part,single label learning is used to collect the mean value of the overall distribution of data.Then,the true value of data label is approached gradually on the granularity of data batch.Finally,the whole network model is retrained using the generated label distribution data.Experimental results show that this method can improve the accuracy of the network model obviously,and has certain competitiveness compared with the advanced algorithms.

Graph Laplacian Matrix Learning from Smooth Time-Vertex Signal

In this paper,we focus on inferring graph Laplacian matrix from the spatiotemporal signal which is defined as“time-vertex signal”.To realize this,we first represent the signals on a joint graph which is the Cartesian product graph of the time-and vertex-graphs.By assuming the signals follow a Gaussian prior distribution on the joint graph,a meaningful representation that promotes the smoothness property of the joint graph signal is derived.Furthermore,by decoupling the joint graph,the graph learning framework is formulated as a joint optimization problem which includes signal denoising,timeand vertex-graphs learning together.Specifically,two algorithms are proposed to solve the optimization problem,where the discrete second-order difference operator with reversed sign(DSODO)in the time domain is used as the time-graph Laplacian operator to recover the signal and infer a vertex-graph in the first algorithm,and the time-graph,as well as the vertex-graph,is estimated by the other algorithm.Experiments on both synthetic and real-world datasets demonstrate that the proposed algorithms can effectively infer meaningful time-and vertex-graphs from noisy and incomplete data.

Wireless distributed computing for cyclostationary feature detection

Recently, wireless distributed computing （WDC） concept has emerged promising manifolds improvements to current wireless technotogies. Despite the various expected benefits of this concept, significant drawbacks were addressed in the open literature. One of WDC key challenges is the impact of wireless channel quality on the load of distributed computations. Therefore, this research investigates the wireless channel impact on WDC performance when the tatter is applied to spectrum sensing in cognitive radio （CR） technology. However, a trade- off is found between accuracy and computational complexity in spectrum sensing approaches. Increasing these approaches accuracy is accompanied by an increase in computational complexity. This results in greater power consumption and processing time. A novel WDC scheme for cyclostationary feature detection spectrum sensing approach is proposed in this paper and thoroughly investigated. The benefits of the proposed scheme are firstly presented. Then, the impact of the wireless channel of the proposed scheme is addressed considering two scenarios. In the first scenario, workload matrices are distributed over the wireless channel

考虑训练样本分布不均衡的超短期风电功率概率预测

提出一种考虑训练样本分布不均衡的超短期风电概率预测方法。首先构建深度信念混合密度网络,通过深度信念网络独特的预训练和微调机制提取输入变量的隐特征,利用Beta混合概率分布的有界性准确表征风电预测功率的概率分布,实现隐特征与预测功率概率分布参数之间的非线性映射;然后引入训练样本分布平滑策略,其中特征分布平滑技术用于校准输入特征,标签分布平滑技术用于对各样本误差赋予差异化权重,从输入和输出两方面改善训练样本分布不均衡现象对预测结果的不利影响。实际算例结果表明,与常见风电功率概率预测模型相比,所提模型在点预测和概率预测方面均能获得较高的预测精度,尤其能有效提高低密度样本区域的预测精度。

基于改进蚁群算法的机器人全局路径规划

针对传统蚁群算法存在初始信息素缺乏、收敛速度慢以及无法有效躲避障碍物等问题,文中提出了一种基于改进蚁群算法的全局路径规划。引入正态分布函数改进传统启发函数,提高了算法效率,缩短了算法收敛所需时间。自适应调整信息素挥发系数,限定信息素范围,避免过早收敛。对算法路径平滑处理,缩短路径长度,从而实现机器人的全局路径规划。仿真结果表明,在20×20环境下,文中算法平均迭代次数比传统蚁群算法减少了28代,收敛速度更快。平均拐点减少了33.3%,使路径更为平滑,克服了初始信息素缺乏,加快了收敛速度,减少了拐点数量,能够有效躲避环境中的障碍物,证明了该算法的可行性。

扩展目标跟踪Student’s t逆Wishart平滑算法

脉冲干扰和离群量测信息等因素通常会导致异常的厚尾噪声,这使得以高斯假设为前提的扩展目标跟踪(ETT)估计器的性能急剧降低,针对该问题该文提出一种基于扩展目标随机矩阵模型(RMM)的Student’s t逆Wishart平滑(StIWS)算法。首先,将目标的运动状态以及过程噪声和量测噪声建模为Student’s t分布以表征异常噪声对扩展目标概率分布的影响,将目标扩展状态建模为服从逆Wishart分布的随机矩阵。然后,在Student’s t贝叶斯平滑框架下,详细推导了能在扩展目标的多重特征动态演变的过程中有效估计目标状态的StIWS算法。最后,通过扩展目标跟踪的仿真实验结果和真实场景实验结果验证了所提算法的有效性。

海上风电多端柔性直流输电及其功率流控制

相比目前已有的工程而言,未来的海上风电场容量将更大,且离岸距离也更远,而基于电压源换流器的多端直流输电系统可以经济有效地联接广域分布的海上风电场,大规模地传输波动的风能至各陆地电网。以大型海上风电场送出为研究对象,提出一种结合电压源换流器和电池储能的多端柔性直流输电系统,给出系统的功率流控制方法,实现网侧换流器的功率分配及功率平滑输出。最后,基于RTDS硬件在环实验平台构建一个四端直流输电测试系统,并用实验验证所提多端柔性直流输电系统功率流的控制效果。

电弧炉异常炉况电极调节应激干预技术的研究

三相交流电弧炉在冶炼过程中常遭遇短路和断弧异常炉况,这不仅延长了通电和出钢时间,增加了能耗,还威胁到生产的安全性和稳定性。提出了一种创新的基于电弧声信号分析的电极调节应激干预技术,以优化EAF冶炼过程。首先基于75t普碳钢EAF生产现场的电弧声信号探究了短路、断弧与稳态运行时三类电弧声时频特征;利用平滑伪维格纳分布算法来评估电弧声的能量密度水平,确定了异常炉况与稳态运行时的特征频率,实现了异常炉况判别。此外,通过分析稳态运行、短路与断弧前期及发生期的电弧声特征频率,揭示了异常炉况的预测规律。在此基础上,在现有的电极调节控制系统中集成了应激干预机制,并通过程序化实现了自动化控制。工程验证结果表明,该技术能够有效地在3~4秒前预测并干预异常炉况,显著减少了短路和断弧事件,同时降低了7.5%的单吨冶炼能耗,证实了其在提升EAF冶炼效率和降低成本方面的潜力。

基于EWT的分布式光储PCC功率波动自适应平抑方法

分布式光伏在交流侧公共连接点(point of common coupling,PCC)汇流的功率有较大的随机性与波动性,影响电网的稳定运行。为此,提出了基于经验小波变换(empirical wavelet transform,EWT)的分布式光储PCC功率自适应平抑方法。首先,针对混合储能(hybrid energy storage system,HESS)与分布式光伏接入PCC的典型场景,在分析EWT自适应处理波形的特点后,结合功率波动率与储能元件的响应特性,对PCC的光伏原始汇流功率进行EWT分解与优化修正,实现HESS的功率初级分配。之后为避免HESS的荷电状态(state of charge,SOC)频繁越限,提出了一种主动功率补偿的SOC控制策略,通过主动改变储能的参考信号使其SOC在安全范围内工作。结合实际数据的仿真验证表明,该平抑方法能够自适应地实现光伏出力的合理分解与功率分配,在延长储能使用寿命的同时有效满足并网功率波动的要求,为平抑光伏输出功率波动提供了新思路。

基于等价输入干扰的分布式光伏电源出力控制策略

分布式电源能够改善配电网潮流分布,使得电力系统具有更高的能源利用率以及灵活性,然而分布式电源的不合理出力则难以充分发挥自身优势,并且死区效应、电网电压畸变及系统参数不确定性等不确定性扰动也会对电力系统的可靠性构成严重威胁。针对此问题,提出了基于改进的等价输入干扰方法(improvedequivalentinputdisturbance,IEID)的分布式电源出力控制方法。其中上层优化结构以配电网网损最小为目标,建立配电网优化降损模型并求解得到分布式电源出力值;下层控制结构基于所提IEID设计分布式电源控制策略,把模型耦合项、死区效应、网压畸变及系统参数不确定性等干扰对系统输出的影响等价到输入通道上,并对其进行估计和补偿,能够有效地抑制未知扰动并高精度跟踪上层输出的出力值,极大地改善并网电能质量。以某21节点真实居民配电网为算例,仿真结果表明所提方法具有较好的扰动抑制能力,能够自主平稳出力,降低配电网网损。

基于RBF神经网络的主动配电网通信过程安全态势感知方法

为提前拦截主动配电网通信过程中的恶意攻击、保证电网稳定运行,提出基于RBF神经网络的通信过程安全态势感知方法。综合考虑电压越限、失负荷、供电能力等因素,筛选主动配电网通信安全态势评价指标;利用非线性基函数实现从输入到输出的非线性变换,采用改进ReliefF算法选取网络通信入侵行为特征,并将其作为RBF神经网络的输入量;将主动配电网络节点部署的安全态势数据分割为若干分量,通过参数调整平滑处理安全态势感知数据,再利用双层滤波检测法输出RBF神经网络的感知结果。由试验结果可知:采用该方法时,特征识别准确率为97.61%,安全态势值为0.894,态势感知时间为0.361 s,能有效感知攻击行为,精准呈现网络态势变化,且感知过程用时较短。

基于计算负荷预测的配用电终端算力供需动态均衡方法

配电网侧海量分布式对象的持续涌入令配用电终端基于确定边界配置转变为柔性资源配置,该转变是在终端有限资源的基础上适应更广泛对象接入的可行保障。为此,提出一种基于计算负荷预测的配用电终端算力供需动态均衡方法,在配用电终端柔性配置转变下最大限度地降低配用电业务与终端供需配合时的计算负担影响,提高终端针对计算负担的平衡能力与稳定能力,实现资源的充分挖掘利用,以适应未来更广泛的对象接入。根据配用电业务的时移特性与调整策略,建立计算负荷模型;基于配用电业务计算负荷曲线的“季节性”特征,构建结合三次指数平滑与迭代马尔可夫链修正的区段预测模型;根据算力资源的交互关系,并结合业务调整策略的划分特点与计算负荷曲线特点,建立配用电业务算力资源均衡模型。算例结果表明:所建预测模型所获区段标准差值满足预测精度要求,所提均衡方法有效降低了算力资源的不平衡度,表现出良好的计算负担削减能力。

基于LADRC的柔性多状态开关平滑切换策略

为实现多端口柔性多状态开关(flexible multi-state switches,FMSS)在所连馈线发生故障时的平滑切换,该文提出一种基于线性自抗扰控制(linear active disturbance rejection control,LADRC)的平滑切换策略,使用LADRC代替比例积分(proportional integral,PI)控制器的直流电压外环并设计解耦状态观测器以便控制器参数调节,引入内环电流指令值反馈以减小其在控制模式切换时的变化幅度,实现FMSS控制模式的平滑切换。通过MATLAB/Simulink仿真,验证了馈线故障出现在PQ或UdcQ控制端口的连接处时,采用基于LADRC的平滑切换策略均可显著减小FMSS的直流电压偏移率,使端口有功功率过渡平滑,迅速恢复稳定,更好地实现了多端口FMSS控制模式的平滑切换。

基于多参数随机变化的高速动车组动力学指标分布研究

针对服役高速动车组关键悬挂部件动力学参数变化后动力学指标分布特征缺乏的问题,对运营里程为145万km的某型动车组的转臂节点径向和轴向刚度、抗蛇行减振器0.01和0.02 m·s^(-1)加载速度下的阻尼力进行测试和统计,结果均服从正态分布;建立并验证动车组参数化的动力学仿真模型,将转臂节点刚度和抗蛇行减振器阻尼力视为随机变量,构建基于改进拉丁超立方抽样的高速动车组动力学仿真方法,包含拉丁超立方抽样、轮轨型面与抽样结果组合、更换动车组仿真模型参数、动车组动力学仿真及动力学指标计算、动力学指标统计分析等内容。采用该方法进行动车组的动力学仿真计算,结果表明:服役近145万km的动车组脱轨系数、轮轴横向力和构架横向加速度随车轮磨耗的增加呈上升趋势,轮重减载率、垂向平稳性、横向平稳性指标呈先上升后下降趋势,但所有指标均未超限;脱轨系数、轮重减载率、轮轴横向力、横向平稳性指标、垂向平稳性指标及构架横向加速度均呈正态分布;脱轨系数分布范围为0.036~0.207,均值为0.075,标准差为0.020;轮重减载率分布范围为0.107~0.262,均值为0.164,标准差为0.036;轮轴横向力分布范围为4.209~16.701 kN,均值为8.438kN,标准差为2.221 kN;横向平稳性指标分布范围为1.351~2.469,均值为1.566,标准差为0.149,垂向平稳性指标分布范围为2.260~2.323,均值为2.273,标准差为0.005;构架横向加速度分布范围为0.906~3.994 m·s^(-2),均值为1.978 m·s^(-2),标准差为0.590 m·s^(-2)。

基于自适应组合窗函数的SPWVD算法及应用研究

针对平滑伪Wigller-Ville分布(Smoothing Pseudo-Winger-Ville distribution,SPWVD)在处理不同信号时,需要采用对应的平滑窗参数才能达到最优时频分布的问题,提出了一种基于自适应组合窗函数的SPWVD算法。该算法将Rényi熵作为时频评价指标,以组合窗函数的窗长和系数作为参数库,提出将随机梯度下降和扰动随机梯度下降分别用于窗长和系数的选取,提升计算速度分别为遍历法的20倍和9倍,最终得到信号的最优时频图。将上述的改进方法应用于碳纤维复合材料的缺陷检测中,结果表明,自适应组合窗函数的SPWVD可以有效地对不同信号选取最优的窗函数参数,对比固定窗函数的SPWVD,自适应组合窗函数的SPWVD具有更好的时频聚集效果,更加完备地对信号进行了表征,并对缺陷进行了定量分析,具备缺陷检测的良好应用前景。

基于发电集群协同划分的分布式光伏发电并网功率调节方法

分布式光伏发电并网的平稳运行至关重要。为了降低功率波动对电网频率的影响,本研究基于发电集群协同划分,设计分布式光伏发电并网功率调节方法。首先,在构建分布式光伏发电模型的基础上,分析其发电输出特性,并跟踪最大输出功率点。然后,基于协调算法动态划分光伏发电集群,并计算无功调节和有功调节范围。最后,在光伏功率波动平抑的基础上,规划并网系统调频储能容量,再构建分布式光伏发电并网功率调节模型,抑制各电池储能系统间功率环流问题,实现功率调节。根据仿真实验可知,在3种典型天气条件下,该方法均能够有效控制并网光伏发电功率。并网系统出现故障后,该方法能够在短时间内快速响应,完成功率平衡调节。

面向机载分布式光电的图像拼接算法

随着航空无人系统技术的迅猛发展,分布式机载图像拼接技术已成为备受瞩目的研究领域。本文针对分布式机载图像拼接中存在的视差大和空间几何变换关系复杂等问题,在APAP图像拼接算法基础上提出了一种改进算法。该算法采用变形处理、线性化单应平滑外推至全局性变换以及网格划分法等方法,有效消除了模糊重影、减少边缘处投影失真,提高算法运行效率。在多个场景下的实验中,该算法表现出更小的对齐误差和更高的图像质量指标,包括均方根误差、峰值信噪比、结构相似度和图像熵等。在进行大规模图像拼接时,该改进算法能够实现154张图像的大规模拼接,得到10 k×10 k的高分辨率全景图像,拼接耗时为138 s。因此,该改进算法具有重要的实际应用价值,可用于分布式机载图像拼接的实际应用中。