This paper presents a small perturbation Cramer method for obtaining the large deviation principle of a family of measures (β,ε> 0) on a topological vector space. As an application, we obtain the moderate deviati...This paper presents a small perturbation Cramer method for obtaining the large deviation principle of a family of measures (β,ε> 0) on a topological vector space. As an application, we obtain the moderate deviation estimations for uniformly ergodic Markov processes.展开更多
In this paper, by considering the stochastic proces s of the busy period and the idle period, and introducing the unfinished work as a supplementary variable, a new vector Markov process was presented to study th e M...In this paper, by considering the stochastic proces s of the busy period and the idle period, and introducing the unfinished work as a supplementary variable, a new vector Markov process was presented to study th e M/G/1 queue again. Through establishing and solving the density evolution equa tions, the busy-period distribution, and the stationary distributions of waitin g time and queue length were obtained. In addition, the stability condition of th is queue system was given by means of an imbedded renewal process.展开更多
随着电网公司代理购电业务稳步推进,代理购电业务体系逐步完善,精确的代理购电用户用电量预测为保障电力安全稳定供应奠定了基础。因此,文章构建自适应权重组合模型,将不同校核方法的校核结果进行权重分配,从而提升校核结果准确性。首先...随着电网公司代理购电业务稳步推进,代理购电业务体系逐步完善,精确的代理购电用户用电量预测为保障电力安全稳定供应奠定了基础。因此,文章构建自适应权重组合模型,将不同校核方法的校核结果进行权重分配,从而提升校核结果准确性。首先,构建预测业务偏差校核流程框架,确定代理购电预测业务校核流程。然后分别选取分位数映射法、增量变化法以及支持向量回归(support vector regression,SVR)对预测结果进行校核,得到同一纬度下的不同方法校核结果。最后,建立遗传算法-优劣解距离法(genetic algorithm-technique for order preference by similarity to ideal solution,GA-TOPSIS)模型针对校核结果进行准确性与稳定性双目标优化,选取不同校核方法的最优权重组合。测试结果表明在校核方法权重组合校正后,相较于初始预测值和单一校核方法校核后的结果,预测精度和准确度得到明显提升。展开更多
文摘This paper presents a small perturbation Cramer method for obtaining the large deviation principle of a family of measures (β,ε> 0) on a topological vector space. As an application, we obtain the moderate deviation estimations for uniformly ergodic Markov processes.
基金Project supported by the National Natural Science Foundation of China(Grant No.70171059)
文摘In this paper, by considering the stochastic proces s of the busy period and the idle period, and introducing the unfinished work as a supplementary variable, a new vector Markov process was presented to study th e M/G/1 queue again. Through establishing and solving the density evolution equa tions, the busy-period distribution, and the stationary distributions of waitin g time and queue length were obtained. In addition, the stability condition of th is queue system was given by means of an imbedded renewal process.
文摘随着电网公司代理购电业务稳步推进,代理购电业务体系逐步完善,精确的代理购电用户用电量预测为保障电力安全稳定供应奠定了基础。因此,文章构建自适应权重组合模型,将不同校核方法的校核结果进行权重分配,从而提升校核结果准确性。首先,构建预测业务偏差校核流程框架,确定代理购电预测业务校核流程。然后分别选取分位数映射法、增量变化法以及支持向量回归(support vector regression,SVR)对预测结果进行校核,得到同一纬度下的不同方法校核结果。最后,建立遗传算法-优劣解距离法(genetic algorithm-technique for order preference by similarity to ideal solution,GA-TOPSIS)模型针对校核结果进行准确性与稳定性双目标优化,选取不同校核方法的最优权重组合。测试结果表明在校核方法权重组合校正后,相较于初始预测值和单一校核方法校核后的结果,预测精度和准确度得到明显提升。
