论文部分内容阅读
为了减少序列离散化过程导致的误差,提出几种概率补偿的方法。由误差的成因入手,分析了离散化过程的误差机理,将利用随机变量的概率分布函数离散化后所得序列,分别以修正公式、平均补偿法及比例补偿法进行补偿,推导出各种误差补偿方法下序列期望值及方差的误差估计公式。结果表明:经误差补偿后,算例中序列期望值误差下降为原误差的8%以下,而方差误差至少可降为原误差的91%。
In order to reduce the error caused by sequence discretization, several methods of probability compensation are proposed. Starting with the cause of the error, the error mechanism of the discretization process is analyzed. The sequence obtained by discretization of the probability distribution function of the random variables is compensated by the correction formula, the average compensation method and the proportional compensation method respectively, and various error compensation Estimation formula of expected value and variance of sequence under the method. The results show that after the error compensation, the error of the expected value in the example is reduced to less than 8% of the original error, and the variance error can be reduced to at least 91% of the original error.