A New Maximum Test via the Dependent Samples t-Test and the Wilcoxon Signed-Ranks Test

A maximum test in lieu of forcing a choice between the two dependent samples t-test and Wilcoxon signed-ranks test is proposed. The maximum test, which requires a new table of critical values, maintains nominal α while guaranteeing the maximum power of the two constituent tests. Critical values, obtained via Monte Carlo methods, are uniformly smaller than the Bonferroni-Dunn adjustment, giving it power superiority when testing for treatment alternatives of shift in location parameter when data are sampled from non-normal distributions.

Modified DS np Chart Using Generalized Multiple Dependent State Sampling under Time Truncated Life Test

This study presents the design of a modified attributed control chart based on a double sampling(DS)np chart applied in combination with generalized multiple dependent state(GMDS)sampling to monitor the mean life of the product based on the time truncated life test employing theWeibull distribution.The control chart developed supports the examination of the mean lifespan variation for a particular product in the process of manufacturing.Three control limit levels are used:the warning control limit,inner control limit,and outer control limit.Together,they enhance the capability for variation detection.A genetic algorithm can be used for optimization during the in-control process,whereby the optimal parameters can be established for the proposed control chart.The control chart performance is assessed using the average run length,while the influence of the model parameters upon the control chart solution is assessed via sensitivity analysis based on an orthogonal experimental design withmultiple linear regression.A comparative study was conducted based on the out-of-control average run length,in which the developed control chart offered greater sensitivity in the detection of process shifts while making use of smaller samples on average than is the case for existing control charts.Finally,to exhibit the utility of the developed control chart,this paper presents its application using simulated data with parameters drawn from the real set of data.

Designing Adaptive Multiple Dependent State Sampling Plan for Accelerated Life Tests

A novel adaptive multiple dependent state sampling plan(AMDSSP)was designed to inspect products from a continuous manufacturing process under the accelerated life test(ALT)using both double sampling plan(DSP)and multiple dependent state sampling plan(MDSSP)concepts.Under accelerated conditions,the lifetime of a product follows the Weibull distribution with a known shape parameter,while the scale parameter can be determined using the acceleration factor(AF).The Arrhenius model is used to estimate AF when the damaging process is temperature-sensitive.An economic design of the proposed sampling plan was also considered for the ALT.A genetic algorithm with nonlinear optimization was used to estimate optimal plan parameters to minimize the average sample number(ASN)and total cost of inspection(TC)under both producer’s and consumer’s risks.Numerical results are presented to support the AMDSSP for the ALT,while performance comparisons between the AMDSSP,the MDSSP and a single sampling plan(SSP)for the ALT are discussed.Results indicated that the AMDSSP was more flexible and efficient for ASN and TC than the MDSSP and SSP plans under accelerated conditions.The AMDSSP also had a higher operating characteristic(OC)curve than both the existing sampling plans.Two real datasets of electronic devices for the ALT at high temperatures demonstrated the practicality and usefulness of the proposed sampling plan.

A Novel Multiple Dependent State Sampling Plan Based on Time Truncated Life Tests Using Mean Lifetime

The design of a new adaptive version of the multiple dependent state(AMDS)sampling plan is presented based on the time truncated life test under the Weibull distribution.We achieved the proposed sampling plan by applying the concept of the double sampling plan and existing multiple dependent state sampling plans.A warning sign for acceptance number was proposed to increase the probability of current lot acceptance.The optimal plan parameters were determined simultaneously with nonlinear optimization problems under the producer’s risk and consumer’s risk.A simulation study was presented to support the proposed sampling plan.A comparison between the proposed and existing sampling plans,namely multiple dependent state(MDS)sampling plans and a modified multiple dependent state(MMDS)sampling plan,was considered under the average sampling number and operating characteristic curve values.In addition,the use of two real datasets demonstrated the practicality and usefulness of the proposed sampling plan.The results indicated that the proposed plan is more flexible and efficient in terms of the average sample number compared to the existing MDS and MMDS sampling plans.

Exponential synchronization of complex dynamical networks with Markovian jumping parameters using sampled-data and mode-dependent probabilistic time-varying delays

In this paper, the problem of exponential synchronization of complex dynamical networks with Markovian jumping parameters using sampled-data and Mode-dependent probabilistic time-varying coupling delays is investigated. The sam- pling period is assumed to be time-varying and bounded. The information of probability distribution of the time-varying delay is considered and transformed into parameter matrices of the transferred complex dynamical network model. Based on the condition, the design method of the desired sampled data controller is proposed. By constructing a new Lyapunov functional with triple integral terms, delay-distribution-dependent exponential synchronization criteria are derived in the form of linear matrix inequalities. Finally, two numerical examples are given to illustrate the effectiveness of the proposed methods.

ASYMPTOTICS OF THE RESIDUALS DENSITY ESTIMATION IN NONPARAMETRIC REGRESSION UNDER m(n)-DEPENDENT SAMPLE

Let Y_i=M(X_i)+ei, where M(x)=E(Y|X=x) is an unknown realfunction on B(? R), {(X_1,Y_i)} is a stationary and m(n)-dependent sample from(X, Y), the residuals {e_i} are independent of {X_i} and have unknown common densityf(x). In [2] a nonparametric estimate f_n(x) for f(x) has been proposed on the basisof the residuals estimates. In this paper, we further obtain the asymptotic normalityand the law of the iterated logarithm of f_n(x) under some suitable conditions. Theseresults together with those in [2] bring the asymptotic theory for the residuals densityestimate in nonparametric regression under m(n)-dependent sample to completion.

Exact Distributions of Waiting Time Problems of Mixed Frequencies and Runs in Markov Dependent Trials

We study waiting time problems for first-order Markov dependent trials via conditional probability generating functions. Our models involve α frequency cells and β run cells with prescribed quotas and an additional γ slack cells without quotas. For any given and , in our Model I we determine the waiting time until at least frequency cells and at least run cells reach their quotas. For any given τ ≤ α + β, in our Model II we determine the waiting time until τ cells reach their quotas. Computer algorithms are developed to calculate the distributions, expectations and standard deviations of the waiting time random variables of the two models. Numerical results demonstrate the efficiency of the algorithms.

基于Electra预训练模型并融合依存关系的中文事件检测模型

事件检测是信息提取领域的一个重要研究方向。现存的事件检测模型受到语言模型训练目标的限制,只能被动地获取词与词之间的依赖关系,使得模型在训练的过程中过多地关注与训练目标不相关的成分,从而导致检测结果错误。以往的研究表明,充分理解上下文信息对于基于深度学习的事件检测技术至关重要。因此,在Electra预训练模型的基础上,引入KVMN网络来捕捉单词之间的依赖关系,以增强单词的语义特征,并采用了一种门控机制来加权这些特征。然后,为了解决中文事件检测中模型识别错误决策的问题,在输入中加入负样本,对不同样本加入不同程度的噪声,使模型学习更好的嵌入表示,有效提高了模型对未知样本的泛化能力。最后,在公共数据集LEVEN上的实验结果表明,该方法优于现有方法,取得了93.43%的F1值。

基于依赖结构和Gibbs Sampling的离散数据聚类

建立了一种新的离散数据聚类方法,该方法结合变量之间的依赖结构和Gibbs sampling进行离散数据聚类,能够显著提高抽样效率,并且避免使用EM算法进行聚类所带来的问题。试验结果表明,该方法能够有效地进行离散数据的聚类。

Novel Stability Criteria for Sampled-Data Systems With Variable Sampling Periods

This paper is concerned with a novel Lyapunovlike functional approach to the stability of sampled-data systems with variable sampling periods. The Lyapunov-like functional has four striking characters compared to usual ones. First, it is time-dependent. Second, it may be discontinuous. Third, not every term of it is required to be positive definite. Fourth, the Lyapunov functional includes not only the state and the sampled state but also the integral of the state. By using a recently reported inequality to estimate the derivative of this Lyapunov functional, a sampled-interval-dependent stability criterion with reduced conservatism is obtained. The stability criterion is further extended to sampled-data systems with polytopic uncertainties. Finally, three examples are given to illustrate the reduced conservatism of the stability criteria.

Asymptotics for partly linear regression with dependent samples and ARCH errors:consistency with rates

Partly linear regression model is useful in practice, but littleis investigated in the literature to adapt it to the real data which are dependent and conditionally heteroscedastic. In this paper, the estimators of the regression components are constructed via local polynomial fitting and the large sample properties are explored. Under certain mild regularities, the conditions are obtained to ensure that the estimators of the nonparametric component and its derivatives are consistent up to the convergence rates which are optimal in the i.i.d. case, and the estimator of the parametric component is root-n consistent with the same rate as for parametric model. The technique adopted in the proof differs from that used and corrects the errors in the reference by Hamilton and Truong under i.i.d. samples.

Statistical inference for dependence competing risks model under middle censoring

Middle censoring is an important censoring scheme,in which the actual failure data of an observation becomes unobservable if it falls into a random interval. This paper considers the statistical analysis of the dependent competing risks model by using the Marshall-Olkin bivariate Weibull(MOBW) distribution.The maximum likelihood estimations(MLEs), midpoint approximation(MPA) estimations and approximate confidence intervals(ACIs) of the unknown parameters are obtained. In addition, the Bayes approach is also considered based on the Gamma-Dirichlet prior of the scale parameters, with the given shape parameter.The acceptance-rejection sampling method is used to obtain the Bayes estimations and construct credible intervals(CIs). Finally,two numerical examples are used to show the performance of the proposed methods.

基于改进PF算法的ADS-B风场反演

将装载自动相关监视广播(ADS-B)机载端的民航飞机视作传感器,利用ADS-B下行数据实施风场反演是近年来新颖的风场探测手段之一。现有的反演算法在飞机小角度转弯情况下存在精度不高的问题,大面积空间风场重建也对风场探测提出了新的挑战。针对上述问题,利用数据的几何结构和Kullback-Leibler Divergence(KLD)采样对粒子滤波(Particle Filter,PF)算法进行改进,该算法通过结合ADS-B数据对航线上的固定位置不同高度层上风矢量进行反演,构造垂直风廓线,并与欧洲中期天气预报中心气象数据作对比,验证了算法的可行性与准确性;最后,利用气象粒子(Meteo-Particle,MP)模型对航线外大面积的风矢量进行估计。实验结果表明,该方法能够反映真实的风场。

考虑预测误差时空相依性的短期风电功率概率预测

由于风电功率点预测误差无法避免,概率预测可以充分描述风电功率的不确定性,进而对调度部门的决策提供进一步指导。当前的风电功率概率预测方法对其物理变化过程挖掘仍不完整。通过挖掘历史风电功率数据和数值天气预报(numerical weather prediction,NWP)的时空特性,构建了一种考虑误差时空相依性的短期风电概率预测新框架。首先通过门控循环单元(gated recurrent unit,GRU)得到点预测结果;进而,引入多位置NWP,提出了一种考虑时空相依特性的多层次误差场景划分方法;最后,利用Bootstrap抽样法重构形成新的适应建模的误差样本集,进行了不同置信水平下的短期风电功率概率预测。由此得出,在考虑时空相依性的概率预测下,整体框架的效果在中国东北某风电场被验证,对比相同置信水平下,预测精度得到有效明显提高,且评价指标预测区间覆盖率提高0.53%、0.44%、0.32%,预测区间平均带宽缩小2.40%、2.14%,0.06%,证实了所提方法的有效性和可行性。

基于马尔科夫毯的近似函数依赖挖掘算法

近似函数依赖挖掘方法通过放宽函数依赖成立条件,允许一定比例的违反,保证原本成立的函数依赖在噪声数据中仍然可以被挖掘出来。然而,现有的发现算法在放宽函数依赖成立条件之后,容易挖掘出大量左部属性数量较多的虚假函数依赖,导致挖掘结果的准确率显著降低。为了解决这一问题,提出基于马尔科夫毯的近似函数依赖挖掘算法,利用马尔科夫毯剪枝左部属性搜索空间,缩小决定项的候选集合,并通过向下泛化算法减少了误差的计算次数,同时降低了复杂度。在保证不丢失真实函数依赖的前提下,避免了近似函数依赖过拟合,从而提高了挖掘结果的准确率。实验结果表明,该方法在真实数据集和合成数据集上的准确率优于现有的近似函数依赖挖掘方法。

结构时变失效概率函数估计的加权重要抽样方法

参数时变失效概率函数(Time-Dependent Failure Probability Function,TDFPF)是结构时变可靠性优化设计的关键问题之一,其估计的精度与效率决定了结构时变可靠性优化的效果。针对时变失效概率函数计算成本高的问题,基于结构复合极限状态思想提出了基于加权重要抽样的结构时变失效概率函数估计策略。对于加权重要抽样进行扩展,仅需一次可靠性分析,即可得到时变失效概率函数的估计,避免了多次重复计算。为了验证所提方法的精度与效率,给出数值算例及工业机器人减速器结构时变算例。结果表明,对于时变失效概率函数估计问题,所提方法适用性好,能够以较高效率获得满意的结果。

基于SVMD和SDRSE方法的输电线路局部放电信号诊断分析

为提高输电线路局部放电信号故障诊断能力,开发一种采用Spearman变分模态分解(SVMD)和空间相关递归样本熵(SDRSE)去除局部放电信号噪声的技术。采用SDRSE可为切比雪夫距离和空间方向评价时间序列不规则程度,构建新型小波阈值去噪函数,并给出信号去噪方法控制流程。仿真信号分析结果表明:采用SVMD-SDRSE方法去噪,形成与初始局部放电(PD)信号相同的效果,可实现优异的抗高斯白噪声,避免周期窄带产生的干扰,获得更优的性能。实测信号分析结果表明:对比其他去噪方法对PD信号去噪的性能,采用该去噪方式可显著抑制最终的信号噪声,达到最大降噪额定值(NRR),对实测信号达到最优去噪。采用SVMDSDRSE方法去噪后,可有效抑制PD信号噪声,具有更优的去噪指标。

ADS-B中基于格的分层无证书消息可恢复认证方案

广播式自动相关监视(ADS-B)作为新一代空中交通管制的关键技术,已部署在全球大部分空域。已有的ADS-B消息认证方案主要利用传统的公钥密码系统来保证数据安全,计算负担较重且无法抵御量子攻击。首次将基于格的密码体制用于ADS-B通信安全,提出一种同时支持消息可恢复和批量验证的分层无证书消息认证方案。ADS-B机载设备不需要管理证书,也无密钥托管问题。ADS-B消息不需要与签名一起传输,可以在验证阶段进行恢复。利用拒绝采样和无陷门技术,该方案仅需执行计算复杂度较低的线性运算即可实现消息认证。在随机预言机模型和小整数解问题假设下,该方案是可证安全的。性能评价实验结果表明,在同等的比特安全级别下,该方案相比相关工作在节省计算开销方面有显著的提升,对于计算资源受限的航空电子设备十分适用。

基于更新样本智能识别算法的自适应集成建模

选择表征建模对象特性漂移的新样本对软测量模型进行自适应更新,能够降低模型复杂度和运行消耗,提高模型可解释性和预测精度.针对新样本近似线性依靠程度(Approximate linear dependence,ALD)和预测误差(Prediction error,PE)等指标只能片面反映建模对象的漂移程度,领域专家结合具体工业过程需要依据上述指标和自身积累经验进行更新样本的有效识别等问题,本文提出了基于更新样本智能识别算法的自适应集成建模策略.首先,基于历史数据离线建立基于改进随机向量泛函连接网络(Improved random vector functional-link networks,IRVFL)的选择性集成模型;然后,基于集成子模型对新样本进行预测输出后采用在线自适应加权算法(On-line adaptive weighting fusion,OLAWF)对集成子模型权重进行更新,实现在线测量阶段对建模对象特性变化的动态自适应;接着基于领域专家知识构建模糊推理模型对新样本相对ALD(Relative ALD,RALD)值和相对PE(Relative PE,RPE)值进行融合,实现更新样本智能识别,构建新的建模样本库;最后实现集成模型的在线自适应更新.采用合成数据仿真验证了所提算法的合理性和有效性.

基于主动学习的中文依存句法分析

目前依存句法分析仍主要采用有指导的机器学习方法,即需要大规模高质量的树库作为训练语料,而现阶段中文依存树库资源相对较少,树库标注又是一件费时费力的工作。面对大量未标注语料,该文将主动学习应用到中文依存句法分析,优先选择句法模型预测不准的实例交由人工标注。该文提出并比较了多种衡量依存句法模型预测可信度的准则。实验表明,一方面,与随机选择标注实例相比,当使用相同数目训练实例时,主动学习使中文依存分析性能最高提升0.8%;另一方面,主动学习使依存分析达到相同准确率时只需标注更少量实例,人工标注量最多可减少30%。