智能测量系统中的数据处理新方法

在智能测量系统中,一般是应用微处理器(MPU)并通过A/D转换器对被测信号进行多次重复采样,然后再对采样数据进行处理.但由于操作者失误、外界突发性强干扰等因素,常会使测量数据产生粗差,含有粗差的数据必须剔除.分析了传统粗差判断准则的原理和不足:当n值有限时,莱特准则和肖维涅准则都不可靠.应用数理统计理论,提出了一种新的粗差判断准则:xi-X>m,并介绍了该方法的实现步骤.经由实际测量系统试验,其更能有效地剔除粗差.

A new data processing method in intelligent measuring system

In the intelligent measuring system, the measured signal is repeatedly acquired by MPU, and the acquired data is preceded. Usually, there is a coarse error in the measured data in the intelligent measuring system because of the operator fault, outer fierce interference, etc. These coarse error data should be eliminated. The principle and fault of the traditional coarse error criterion are introduced(the traditional coarse error criterion are unreliable when the value of n is limited).Based on statistical estimation theory, a new coarse error criterion is shaped and the application of method introduced.The expected result is achieved in the experiment of actual measuring system.