Realization of a streamlined soft-sensing method for leading variables in the production process by means of pearson correlation coefficient

Abstract

The invention discloses a production process dominant variable simplified soft measuring method achieved with help of a Pearson correlation coefficient. The production process dominant variable simplified soft measuring method achieved with the help of the Pearson correlation coefficient is characterized by comprising the following steps: 1 determining n primary auxiliary variables which are correlative with a dominant variable, collecting value data of the n primary auxiliary variables and the dominant variable and enabling the value data of the n primary auxiliary variables and dominant variable to be combined into a sample set; 2 utilizing the Pearson correlation coefficient algorithm to respectively calculate weighted values of the n primary auxiliary variables; 3 forming a primary auxiliary variable sequence; 4 conducting modeling and determining an optimum auxiliary variable according to a minimum mean square error (MSE); and 5 obtaining a simplified soft measuring model. The production process dominant variable simplified soft measuring method can find out an auxiliary variable set with the least number of auxiliary variables to conduct modeling to the dominant variable, and therefore a simplified soft measure to the dominant variable is achieved.

Description

technical field [0001] The invention belongs to the technical field of soft measurement, and in particular relates to a method for realizing simplified soft measurement of leading variables in a production process by means of a Pearson correlation coefficient. Background technique [0002] So far, in the actual production process, there are many variables that cannot be directly measured due to technical or economic reasons. In this case, soft sensor technology emerges as the times require. Soft sensing is based on the mathematical relationship between measurable and easy-to-measure process variables (called auxiliary variables) and difficult-to-detect directly measured variables (called leading variables), according to a certain optimal criterion, using various calculation methods, using The software realizes the measurement or estimation of the measured variable. Soft sensor technology is a hot spot in current research. For example, Chinese patent (patent number: 20041001...

AI Technical Summary

Problems solved by technology

Computing many auxiliary variables to realize the soft measurement of the leading variable will bring a huge amount of calculation, which is not only time-consuming and labor-intensive, but also the obtained soft measurement results are not necessarily the best, which is in the generation process unwanted things

Images