论文部分内容阅读
首先根据三个钢种的大子样试验数据,论证了却贝(Cv)、动态撕裂(DT)试验在任一指定温度下的冲击韧性(能量)服从正态分布。然后,引用了概率统计中的百分误差分析公式,制出了相应的分析用曲线。由这两个方面构成了确定试样个数的理论基础。并据此基础,对多种金属材料的多批次试验数据的百分误差作了统计推断。 最后指出,在80％可信度意义上,两个DT试样均值的百分误差相当于三个Cv试样的均值百分误差。据此,对于国外某些材料技术条件中只取一个DT试样的问题进行了讨论,提出了改进意见。
First of all, based on the large sub-sample data of the three steels, it is demonstrated that the impact toughness (energy) of the Cv and DT test at any given temperature follows a normal distribution. Then, the percentage error analysis formula in probability statistics is quoted, and the corresponding analysis curve is made. These two aspects constitute the theoretical basis for determining the number of samples. Based on this, a statistical inference was made on the percentage error of multiple batch test data of various metal materials. Finally, it is pointed out that in the sense of reliability of 80%, the percentage error of the mean value of two DT samples is equivalent to the mean percentage error of three Cv samples. Based on this, we discussed the problem that only one DT sample was taken from the technical conditions of some foreign materials and put forward the suggestions for improvement.