Alloy Grade Identification Method Based on Pearson Correlation Coefficient

Abstract

The invention discloses a method for identifying alloy grades based on pearson correlation coefficient. The method is as follows: (1) determining a matching element set; (2) testing the element content information of a sample to be identified by a spectrometer, and extracting the content value X of each element in the matching element set i ; (3) Extract the content upper limit X of each element in a certain brand in the matched element set maxi value, lower limit X mini value; (4) Construct the actual matching vector V real and the ideal matching vector V ideal : (5) Calculate the actual matching vector V real and the ideal matching vector V ideal (6) Repeat steps (3)-(5) until the pearson correlation coefficients of all grades in the grade library are calculated, and the grade with the largest correlation coefficient is the grade of the sample to be identified. The calculation process of this method is simple and the calculation result is normalized data, dimensionless, its value ranges from -1 to 1, a value of 1 indicates that the brand is completely consistent with the measurement result, and in other cases, the closer the value is to 1, the match Higher degrees and values ​​closer to ‑1 indicate a lower degree of match.

Description

technical field [0001] The invention relates to the field of alloy detection, in particular to an alloy grade identification method based on Pearson correlation coefficient. Background technique [0002] Brand identification refers to comparing the measured element content information of the tested sample with the element content range information of several brands in the brand library, and inferring related information such as the brand name of the test object. Brand identification is a new function gradually derived with the development of the alloy testing industry, which greatly facilitates the identification and classification of alloys. [0003] There are few types of existing brand identification methods, which are generally divided into the following categories: (1) Determine whether the detected object is the brand according to whether the measured element content value is within the content range specified by the brand. The disadvantage of this method is that if a ...

AI Technical Summary

Problems solved by technology

[0005] In view of the above deficiencies, the present invention provides a method for identifying alloy grades based on pearson correlation coefficient, which introduces the correlation coefficient into the problem of grade recognition, which solves the problems of complex calculation and difficult maintenance of traditional algorithms. What is more prominent is that the method not only Applicable to grade identification problems that provide element content ranges and also applicable to grade identification problems that only provide element content reference values

Images