论文部分内容阅读
函数的误差,由各直接测量量的误差与各误差间的相关性共同决定。结合试验研究工作的实际,本文指出:如函数只包括二、三个变量,则对这二、三个变量应力求做到能各有十几个测量数据;如函数包括四个以上的变量,这四个变最的测量数据各自可以只有几个;在这两种情况下,均可近似按正态分布报导测试结果。在报导函数的测试结果时,通常取置信概率ξ=68.3％,信置系数T_ξ=1。用函数的误差公式,还可以指导测试工作中对所用仪器的合理选配。
The error of the function is determined by the correlation between the error of each direct measurement and each error. Combined with the experimental work, this paper points out: if the function only includes two or three variables, we should strive to achieve a dozen or so of these two or three variables; if the function includes more than four variables, The four most varied measurements can each have only a few; in both cases, the test results can be reported approximately in a normal distribution. When reporting the test result of the function, the probability of taking the letter ξ = 68.3% is usually taken into account, and the signal setting coefficient T_ξ = 1. With the function of the error formula, you can also guide the test of the rational use of equipment matching.