提出一种基于粒子概率假设密度滤波器(Sequential Monte Carlo probability hypothesis density filter,SMC-PHDF)的部分可分辨的群目标跟踪算法.该算法可直接获得群而非个体的个数和状态估计.这里群的状态包括群的质心状态和形状.为了...提出一种基于粒子概率假设密度滤波器(Sequential Monte Carlo probability hypothesis density filter,SMC-PHDF)的部分可分辨的群目标跟踪算法.该算法可直接获得群而非个体的个数和状态估计.这里群的状态包括群的质心状态和形状.为了估计群的个数和状态,该算法利用高斯混合模型(Gaussian mixture models,GMM)拟合SMC-PHDF中经重采样后的粒子分布,这里混合模型的元素个数和参数分别对应于群的个数和状态.期望最大化(Expectation maximum,EM)算法和马尔科夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)算法分别被用于估计混合模型的参数.混合模型的元素个数可通过删除、合并及分裂算法得到.100次蒙特卡洛(Monte Carlo,MC)仿真实验表明该算法可有效跟踪部分可分辨的群目标.相比EM算法,MCMC算法能够更好地提取群的个数和状态,但它的计算量要大于EM算法.展开更多
Although Markov chain Monte Carlo（MCMC） algorithms are accurate, many factors may cause instability when they are utilized in reliability analysis; such instability makes these algorithms unsuitable for widespread e...Although Markov chain Monte Carlo（MCMC） algorithms are accurate, many factors may cause instability when they are utilized in reliability analysis; such instability makes these algorithms unsuitable for widespread engineering applications. Thus, a reliability modeling and assessment solution aimed at small-sample data of numerical control（NC） machine tools is proposed on the basis of Bayes theories. An expert-judgment process of fusing multi-source prior information is developed to obtain the Weibull parameters＇ prior distributions and reduce the subjective bias of usual expert-judgment methods. The grid approximation method is applied to two-parameter Weibull distribution to derive the formulas for the parameters＇ posterior distributions and solve the calculation difficulty of high-dimensional integration. The method is then applied to the real data of a type of NC machine tool to implement a reliability assessment and obtain the mean time between failures（MTBF）. The relative error of the proposed method is 5.8020×10-4 compared with the MTBF obtained by the MCMC algorithm. This result indicates that the proposed method is as accurate as MCMC. The newly developed solution for reliability modeling and assessment of NC machine tools under small-sample data is easy, practical, and highly suitable for widespread application in the engineering field; in addition, the solution does not reduce accuracy.展开更多
居民用电所占比例逐渐提高,对配电网影响日益增大。有效的家庭日负荷曲线模型对需求侧管理及智能电网技术的发展至关重要。该文建立了基于用户行为的家庭日负荷曲线模型。基于测量数据,建立典型居民负荷电气学模型;基于统计调研数据,利...居民用电所占比例逐渐提高,对配电网影响日益增大。有效的家庭日负荷曲线模型对需求侧管理及智能电网技术的发展至关重要。该文建立了基于用户行为的家庭日负荷曲线模型。基于测量数据,建立典型居民负荷电气学模型;基于统计调研数据,利用马尔科夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)算法,引入概率函数表示居民人口、家用电器拥有情况等居民家庭特征的影响,建立居民负荷行为学模型。并采用自下向上的分层建模思路,结合电气学模型与行为学模型建立家庭日负荷曲线模型,同时搭建了仿真平台。所建模型具有系统性和通用性,仿真与实测对比分析验证了该文所提模型的可行性与准确性。展开更多
海底金矿矿山水害对矿山生产、人员施工及矿山设备等产生较大威胁,是矿山开采中的自然灾害之一,快速有效的判别出矿山水害水源对于事故的防治有重要意义.三山岛金矿的巷道围岩裂隙普遍并长期存在涌水现象,矿区开采中矿井水害的水源主要...海底金矿矿山水害对矿山生产、人员施工及矿山设备等产生较大威胁,是矿山开采中的自然灾害之一,快速有效的判别出矿山水害水源对于事故的防治有重要意义.三山岛金矿的巷道围岩裂隙普遍并长期存在涌水现象,矿区开采中矿井水害的水源主要有海水、第四系水、基岩裂隙水、地下水等,为了准确快速的判别矿井水水源,有效预防矿井水突水及水害威胁,本研究结合监测点水样的水文地质条件与不同监测点水样的水化学成分分析,选取Mg^2+、Na^++K^+、Ca^2+、SO4^2-、Cl^-和HCO3^-共6项指标作为判别因子,通过主成分分析得出不同水样的矿化程度.在贝叶斯算法分析原理的基础上,将马尔可夫链蒙特卡洛(Markov Chain Monte Carlo, MCMC)引入到贝叶斯方法中,运用统计软件SPSS统计,构建贝叶斯判别分析模型,得出基于水样样本信息的算法估计的后验分布,得出矿山水害水源的分析方法.运用三山岛金矿水害取水点的水样分析数据进行详细的分析验证,建立矿井突水水源模型,进行不同水样的信息分析,得出贝叶斯统计函数并进行水源判别结果分析,验证了贝叶斯矿山水害水源判别模型的准确性和实用性,对现场工作的开展和水害防治有一定的指导意义.展开更多
针对环境中障碍物对被跟踪目标构成不可预知的遮挡问题,提出了一种新的基于局部区域特征匹配的跟踪算法。首先采用一组基本观察片图模拟目标的外观;其次提出了一种将运动轨迹特性与动态模型结合的采样结构,采用马尔可夫链蒙特卡洛(MCMC,...针对环境中障碍物对被跟踪目标构成不可预知的遮挡问题,提出了一种新的基于局部区域特征匹配的跟踪算法。首先采用一组基本观察片图模拟目标的外观;其次提出了一种将运动轨迹特性与动态模型结合的采样结构,采用马尔可夫链蒙特卡洛(MCMC,Markov chain Monte Carlo)方法独立估计每个基本片图的状态,并使用运动区域一致性规则选择构成目标的有效的特征片图,遮挡状态则被定义为对应片图的消失;最后由有效片图的组合确定目标的可见概率。实验结果表明,与基于单一区域的方法和基于空间互连的多区域方法相比,本文提出的方法在部分或全部遮挡情况下能够更有效预测被跟踪目标的状态。展开更多
This paper investigates Bayesian methods for aerospace system reliability analysis using various sources of test data and expert knowledge at both subsystem and system levels. Four sce- narios based on available infor...This paper investigates Bayesian methods for aerospace system reliability analysis using various sources of test data and expert knowledge at both subsystem and system levels. Four sce- narios based on available information for the priors and test data of a system and/or subsystems are studied using specific Bayesian inference techniques. This paper proposes the Bayesian melding method for integrating subsystem-level priors with system-level priors for both system- and subsystem-level reliability analysis. System and subsystem reliability outcomes are compared under different scenarios. Computational challenges for posterior inferences using the sophisticated Bayesian melding method are addressed using Markov Chain Monte Carlo （MCMC） and adaptive Sam- piing Importance Re-sampling （SIR） methods. A case study with simulation results illustrates the applications of the proposed methods and provides insights for aerospace system reliability analysis using available multilevel information.展开更多
文摘提出一种基于粒子概率假设密度滤波器(Sequential Monte Carlo probability hypothesis density filter,SMC-PHDF)的部分可分辨的群目标跟踪算法.该算法可直接获得群而非个体的个数和状态估计.这里群的状态包括群的质心状态和形状.为了估计群的个数和状态,该算法利用高斯混合模型(Gaussian mixture models,GMM)拟合SMC-PHDF中经重采样后的粒子分布,这里混合模型的元素个数和参数分别对应于群的个数和状态.期望最大化(Expectation maximum,EM)算法和马尔科夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)算法分别被用于估计混合模型的参数.混合模型的元素个数可通过删除、合并及分裂算法得到.100次蒙特卡洛(Monte Carlo,MC)仿真实验表明该算法可有效跟踪部分可分辨的群目标.相比EM算法,MCMC算法能够更好地提取群的个数和状态,但它的计算量要大于EM算法.
基金Supported by Research on Reliability Assessment and Test Methods of Heavy Machine Tools,China(State Key Science&Technology Project High-grade NC Machine Tools and Basic Manufacturing Equipment,Grant No.2014ZX04014-011)Reliability Modeling of Machining Centers Considering the Cutting Loads,China(Science&Technology Development Plan for Jilin Province,Grant No.3D513S292414)Graduate Innovation Fund of Jilin University,China(Grant No.2014053)
文摘Although Markov chain Monte Carlo（MCMC） algorithms are accurate, many factors may cause instability when they are utilized in reliability analysis; such instability makes these algorithms unsuitable for widespread engineering applications. Thus, a reliability modeling and assessment solution aimed at small-sample data of numerical control（NC） machine tools is proposed on the basis of Bayes theories. An expert-judgment process of fusing multi-source prior information is developed to obtain the Weibull parameters＇ prior distributions and reduce the subjective bias of usual expert-judgment methods. The grid approximation method is applied to two-parameter Weibull distribution to derive the formulas for the parameters＇ posterior distributions and solve the calculation difficulty of high-dimensional integration. The method is then applied to the real data of a type of NC machine tool to implement a reliability assessment and obtain the mean time between failures（MTBF）. The relative error of the proposed method is 5.8020×10-4 compared with the MTBF obtained by the MCMC algorithm. This result indicates that the proposed method is as accurate as MCMC. The newly developed solution for reliability modeling and assessment of NC machine tools under small-sample data is easy, practical, and highly suitable for widespread application in the engineering field; in addition, the solution does not reduce accuracy.
文摘居民用电所占比例逐渐提高,对配电网影响日益增大。有效的家庭日负荷曲线模型对需求侧管理及智能电网技术的发展至关重要。该文建立了基于用户行为的家庭日负荷曲线模型。基于测量数据,建立典型居民负荷电气学模型;基于统计调研数据,利用马尔科夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)算法,引入概率函数表示居民人口、家用电器拥有情况等居民家庭特征的影响,建立居民负荷行为学模型。并采用自下向上的分层建模思路,结合电气学模型与行为学模型建立家庭日负荷曲线模型,同时搭建了仿真平台。所建模型具有系统性和通用性,仿真与实测对比分析验证了该文所提模型的可行性与准确性。
文摘海底金矿矿山水害对矿山生产、人员施工及矿山设备等产生较大威胁,是矿山开采中的自然灾害之一,快速有效的判别出矿山水害水源对于事故的防治有重要意义.三山岛金矿的巷道围岩裂隙普遍并长期存在涌水现象,矿区开采中矿井水害的水源主要有海水、第四系水、基岩裂隙水、地下水等,为了准确快速的判别矿井水水源,有效预防矿井水突水及水害威胁,本研究结合监测点水样的水文地质条件与不同监测点水样的水化学成分分析,选取Mg^2+、Na^++K^+、Ca^2+、SO4^2-、Cl^-和HCO3^-共6项指标作为判别因子,通过主成分分析得出不同水样的矿化程度.在贝叶斯算法分析原理的基础上,将马尔可夫链蒙特卡洛(Markov Chain Monte Carlo, MCMC)引入到贝叶斯方法中,运用统计软件SPSS统计,构建贝叶斯判别分析模型,得出基于水样样本信息的算法估计的后验分布,得出矿山水害水源的分析方法.运用三山岛金矿水害取水点的水样分析数据进行详细的分析验证,建立矿井突水水源模型,进行不同水样的信息分析,得出贝叶斯统计函数并进行水源判别结果分析,验证了贝叶斯矿山水害水源判别模型的准确性和实用性,对现场工作的开展和水害防治有一定的指导意义.
文摘针对环境中障碍物对被跟踪目标构成不可预知的遮挡问题,提出了一种新的基于局部区域特征匹配的跟踪算法。首先采用一组基本观察片图模拟目标的外观;其次提出了一种将运动轨迹特性与动态模型结合的采样结构,采用马尔可夫链蒙特卡洛(MCMC,Markov chain Monte Carlo)方法独立估计每个基本片图的状态,并使用运动区域一致性规则选择构成目标的有效的特征片图,遮挡状态则被定义为对应片图的消失;最后由有效片图的组合确定目标的可见概率。实验结果表明,与基于单一区域的方法和基于空间互连的多区域方法相比,本文提出的方法在部分或全部遮挡情况下能够更有效预测被跟踪目标的状态。
文摘This paper investigates Bayesian methods for aerospace system reliability analysis using various sources of test data and expert knowledge at both subsystem and system levels. Four sce- narios based on available information for the priors and test data of a system and/or subsystems are studied using specific Bayesian inference techniques. This paper proposes the Bayesian melding method for integrating subsystem-level priors with system-level priors for both system- and subsystem-level reliability analysis. System and subsystem reliability outcomes are compared under different scenarios. Computational challenges for posterior inferences using the sophisticated Bayesian melding method are addressed using Markov Chain Monte Carlo （MCMC） and adaptive Sam- piing Importance Re-sampling （SIR） methods. A case study with simulation results illustrates the applications of the proposed methods and provides insights for aerospace system reliability analysis using available multilevel information.
