Statistical Inference Methods Of Material Strength With The Consideration Of Measurement Uncertainty

Rebound Method which own simply operation is the main method of concrete compressive strength test. This method is widely used, but there are also inadequacies during the test. First, the statistical inference of concrete compressive strength did not consider the impact of uncertainty in measurement. Second, in the current inference methods, the large and small samples have a lower standard, which lead to a greater deviation. Third, in the actual process of inference, there may have a phenomenon which minimum value of sample is less than inferred, which can not be considered fully in current inference methods. Because of those inadequacies, this paper has a deeply research, the main work and achievements are as follows:(1)This paper reveals system error and uncertainty of rebound method through statistical analysis of 250 standard dates of contrast experiment of "Rebound method—Test block strength", and a new concept of uncertainty is introduced to the nondestructive testing area for the first time.(2)Inference model in current specification only applies to the large sample statistical inference. Through comparative analysis, the sample capacity of 30 is suggested as the line between large sample and small sample. The method of interval estimation should be taken when sample capacity less than 30.(3)Inference method which takes minimum value of sample considering measurement uncertainty was proposed, which can be turn out to be reasonable though analysis of the rate guarantee.(4) Inference method based on sample's statistical results and the minimum value of sample was proposed in this paper.The achievements of this paper are the supplement and improvement of Technical Specification for Inspection of Concrete Compressive Strength by Rebound Method (JGJ/T 23-2001), which can provide reference for later specification amendment, and the basic idea of this paper is also applicable to study other nondestructive testing technologies and evaluation method.