This year, there are more stand-alone performance tests for instruments in surface water, and I also see a little more about the indicators of TOC instrument performance tests. In the TOC water quality automatic analyzer technical requirements (HJ/T 104-2003), there is an indicator of "repeatability error", which stipulates that this "repeatability error" instrument must be within the range of ±5%. The test method is to measure the zero point calibration solution 6 times between 0-40℃ and the temperature change within ±5℃/d. The average value of each indicated value is used as the zero point value. Under the same conditions, measure the span calibration fluid (ie 80% range standard solution) 6 times, calculate the relative standard deviation of the measured value of the 6 times range.

How to calculate the relative standard deviation, as shown in the figure below

Accuracy: the degree to which the measured value matches the true value

        Absolute error: The difference between the measured value (or the average value of multiple determinations) and the true (real) value is called the absolute error, expressed by δ, and the absolute error can be positive or negative.

        Relative error: The ratio of absolute error to true value is called relative error. It is often expressed as a percentage. It can indicate the accuracy of the measuring instrument, but it cannot reflect the proportion of the error in the measurement value. The relative error reflects the proportion of the measurement error in the measurement result. It is more meaningful to measure the relative error.

        True value (μ): The true value exists objectively, but there are errors in any measurement, so the true value can only be approximated but not measurable. In actual work, the “standard value” is often used instead of the “true value”. In the stand-alone performance test, we use the standard solution prepared by the laboratory analyst as the true value.

        Precision: The degree to which the results of several parallel determinations are close to each other. The closer the results of each measurement are, the higher the precision, and the deviation is used to measure the precision.

        Deviation: The difference between a single measurement value and the sample average value: