In the detection work, detection error, detection accuracy, and detection uncertainty are frequently used terms, which are directly related to the reliability of the detection results and the accuracy and consistency of the value. It is easy to confuse or mistake the three in daily work. This article will compare the definitions and differences of the three so that they can be easily distinguished at work.

Definition of three numbers

The detection error refers to the detection result minus the detected true value, which is the difference between the detection result and the detected true value. The so-called true value refers to the true size of the quantity itself when detecting a quantity. The true value of a quantity is an ideal concept, which is generally unknown. The value measured by the standard of accuracy is called the actual value, and the actual value is often used instead of the true value. In some specific cases, the true value is known. For example, the circumferential angle of a full circle is 360°, and the sum of the three interior angles of a triangle is 180°. According to the definition, the value of the international kilogram reference can be regarded as the true value of 1kg. Wait.

Detection accuracy refers to the degree of agreement between the detection result and the true value being detected. Since the exact magnitude of the true value cannot be known in many cases, accuracy is defined as the closeness of the detection result to the detected true value. It is a qualitative concept and cannot be operated as a quantity, which means that it can be said that the accuracy is high or low, the accuracy is 0.2, the accuracy is 3, etc. and the accuracy meets the XX standard, etc. The technical indicators of a certain grade or level require or meet the requirements of a technical specification.

Detection uncertainty refers to a parameter that characterizes the dispersion of the value reasonably assigned to the detected value and is associated with the detection result. It represents the dispersion of the detected value, and represents an interval, that is, the interval in which the detected value may be distributed. According to its acquisition method, it is divided into two types of evaluation components, A and B. The evaluation component of type A is made through statistical analysis of observation columns. Uncertainty ratings, Category B rating components are estimated based on experience or insight, and assume that there are uncertainty components characterized by an approximate standard deviation.

The difference between the three values

By understanding the definitions of the three values, it can be seen that the detection error, detection accuracy and detection uncertainty are different in the following aspects.

1. The influencing factors of the three values are different

The detection error exists objectively, is not affected by external factors, and does not change according to the degree of human cognition. Any detection has its imperfections, so errors will occur at any time; the detection uncertainty is analyzed and evaluated by people. obtained, so it is related to people's understanding of the detected, affected quantity and the detection process; the detection accuracy is the level of the precision of the instrument used in the detection process, and the level is related to the level. The higher the precision level, the better its accuracy. The closer the measurement result is to the true value.

2. The purpose of the three numerical evaluations is different

The detection error is the difference between the detection result and the true value, in order to indicate the degree to which the detection result deviates from the true value, it is a quantitative concept, and the specific deviation is how much; the detection accuracy is the consistency between the detection result and the detected true value Degree, it is a qualitative concept, indicating whether the measured results are within the standard requirements; the detection uncertainty is an estimate of the dispersion of the impact error, in order to indicate the dispersion of the detected value, indicating that the detected The result is within a certain interval.

3. The results of the three numerical evaluations are different

The detection error is a value with positive and negative signs, which is the specific value obtained by each test, which can only be obtained by testing. It is a point on the number axis; the detection uncertainty represents an interval, that is, the detected value may be The range of the distribution, which represents a segment (interval) on the number line, is an unsigned parameter and is represented by the standard deviation or the multiple of the standard deviation or the half-width of the confidence interval. It is evaluated by people based on experimental data, experience and other information. It is quantitatively determined by the two types of evaluation methods A and B; the detection accuracy is the accuracy of the detection equipment. It is not a quantity and cannot be calculated as a quantity. It indicates the accuracy of the detection results.

4. The three values are generated for different reasons

The reason for the detection error is the defect of the detection process, which is caused by a variety of reasons, including the error generated by the detection device, the error generated by the environmental conditions, the error generated by the detection method, the error generated by the personnel, etc. It can be divided into systematic errors. , random error and gross error; detection accuracy is the result affected by systematic errors, such as the error generated by the detection device, which determines the accuracy of the detection result; the cause of detection uncertainty and detection principle, detection instrument, detection environment Conditions, testing procedures, testing personnel and data processing methods, etc., as well as the experience and knowledge scope of the assessor are related to the understanding of water evaluation. It consists of multiple components, some of which can be estimated by the statistical distribution of the testing results, and the experimental standard deviation. Other components can be estimated using assumed probability distributions based on experimental or other information, and can also be characterized by standard deviations.

The connection between the three values

We can see that the detection error, detection accuracy and detection uncertainty are different from the above, but they are still closely related. The concept of uncertainty is the application and extension of error theory, while error analysis is the evaluation of detection uncertainty. The theoretical basis for estimating class B components is inseparable from error analysis; the detection uncertainty is mostly used as a technical indicator of the testing instrument, which indicates the response ability of the measured result close to the true value.

Email: hello@utstesters.com

Direct: + 86 152 6060 5085

Tel: +86-596-7686689

Web: www.utstesters.com