Method for screening sensitive variables influencing product quality in complex industrial process based on step-by-step reduction

Abstract

The invention discloses a method for screening sensitive variables influencing product quality in a complex industrial process based on step-by-step reduction, which belongs to the technical field ofsoft measurement and comprises the following steps of: selecting auxiliary variables influencing the product quality through expert knowledge and collecting data samples; calculating an auxiliary variable sensitivity index by comprehensively considering the variable correlation and the sensitivity of the variable to the working condition change, and preliminarily screening the sensitive variable influencing the dominant variable; and constructing a weighted cosine Mahalanobis system, and accurately screening key sensitive variables influencing the product quality. According to the method, thecorrelation and the working condition information of the variables can be accurately reflected, meanwhile, the redundancy of the variables is well reduced, the product quality prediction precision canbe improved, the complexity of the prediction model can be effectively reduced, and the method also has important significance for maintenance of the soft sensor model.

Description

technical field [0001] The invention relates to the technical field of soft measurement, in particular to a method for screening sensitive variables affecting product quality based on step-by-step reduction. Background technique [0002] With the development of advanced manufacturing technology, the manufacturing industry has put forward higher requirements for production development from the expansion of quantity and scale to the improvement of quality, efficiency and environmental protection. In order to be able to monitor and evaluate the process operation status in a timely and effective manner, to realize accurate diagnosis of system faults and fast tracking of product quality, it is necessary to conduct real-time detection of process-critical product quality. However, due to the harshness of the testing environment, the high cost of analytical instruments, and the hysteresis of laboratory analysis, it is currently difficult to achieve online testing for the quality of ...

AI Technical Summary

Problems solved by technology

The filtering variable selection method does not depend on the learning algorithm, and adapts the learning algorithm by changing the data, but this method tends to ignore the variable correlation, resulting in the selected subset may not be the optimal subset

[0005] In order to solve the problem of variable redundancy in the filtering variable selection method, many attempts and researches have been carried out in the theoretical circles and industrial practices at home and abroad. These studies can effectively solve the problem that the filtering variable selection method tends to ignore the correlation and redundancy between variables. , but does not have the ability to describe the process working condition information

However, in the actual industrial production process, affected by fluctuations in the quality of imported raw materials, adjustments in processing schemes, and changes in product specification requirements, the production conditions are in a state of fluctuation, and there will be certain differences in product quality under different conditions.

If the selected auxiliary variables cannot describe the changes of working conditions well, the accuracy of the prediction model will be reduced to a certain extent.

Images