Visual quality grading method and system for multi-index industrial products

Abstract

The invention discloses a visual quality grading method and system for multi-index industrial products and belongs to the field of quality monitoring of industrial products. According to the method, amodeling method based on multivariate statistical analysis is adopted, a direction with a large multivariate data variance is selected to represent the correlation information, division is carried out to obtain the principal component space and the residual error space, statistics are respectively constructed for monitoring, the correlation information among different indexes can be brought intoa detection range under the condition that the number of measurement indexes is large, and grading standards can be accurately and effectively formulated and the product quality can be monitored for the multi-index industrial product. According to the method, the grading boundary is determined by using the Bayesian diagram, the main correlation information can be reserved, the method is intuitiveand efficient, and quality monitoring efficiency is improved; distribution characteristics of each index are extracted by utilizing a maximum likelihood estimation method, and a large number of randomsamples are generated according to the distribution characteristics, so possible quality grading range loss under the condition that the number of historical data is relatively small is avoided, andgrading accuracy is further ensured.

Description

technical field [0001] The invention belongs to the field of quality monitoring of industrial products, and more specifically relates to a visual quality grading method and system for multi-indicator industrial products. Background technique [0002] Modern industrial systems tend to be complex and refined, and the quality of their products may have a direct impact on economic benefits. Based on user needs and actual characteristics, the quality of a product often needs to be reflected by multiple indicators. It is of great significance to reasonably combine multiple detection indicators to accurately and intuitively classify the quality of products. [0003] The currently widely used quality grading methods are mainly methods based on univariate monitoring, which formulate grade boundaries for different indicators. In practical applications, there may be multiple indicators that are correlated with each other, that is, there is a correlation between the indicators. At this...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the above defects or improvement needs of the prior art, the present invention provides a visual quality grading method and system for multi-indicator industrial products. The boundary is unreasonable and the accuracy is insufficient; due to the need to monitor each indicator separately, when the number of detection indicators is large, there is a problem that it is not intuitive enough

Images