Layered integrated Gaussian process regression soft measurement modeling method

Abstract

The invention discloses a layered integrated Gaussian process regression soft measurement modeling method used for a complex changeable multi-stage chemical process. The layered integrated Gaussian process regression soft measurement modeling method is an on-line multi-model strategy. A Gaussian mixture model is employed to identify different stages of the process, principal component analysis is carried out on data in each stage, on the basis of the contribution degree of each auxiliary variable in the principal element space, data in each mode is divided into several subspaces, and a corresponding Gaussian process regression soft measurement model is established. When new data comes around, variable selection is carried out by means of subspace PCA, and on the basis of the soft measurement model which is established off line, the prediction output of each model can be obtained. By carrying out mean value fusion on outputs of subspace models, first layer integrated output, i.e., local prediction output in each mode can be obtained, finally new data obtained according to calculation is attached to the posterior probability of each different stage, and local prediction in each mode is fused by means of the posterior probability to obtain second layer integrated output. Key variables can be accurately predicted, and therefore the product quality is improved, and the production cost is reduced.

Description

technical field [0001] The invention relates to a layered and integrated Gaussian process regression soft sensor modeling method, which belongs to the field of complex industrial process modeling and soft sensor. Background technique [0002] Some important quality variables in industrial processes such as chemical industry, metallurgy and fermentation are often impossible or difficult to measure through online instruments. Based on the data-driven soft-sensing modeling method, which does not require in-depth knowledge of the mechanism of the process, it has been widely used in industrial process modeling. Commonly used linear modeling methods such as principal component regression (Principal component regression, PCR) and partial least squares (Partial least squares, PLS), etc., can handle and model the linear relationship between process data well. [0003] However, the chemical process often presents significant nonlinear features, so nonlinear modeling methods such as a...

AI Technical Summary

Problems solved by technology

Conventional ensemble learning only builds a soft sensor model from a single sample or variable dimension, and does not make full use of the two-dimensional information contained in the sample. Therefore, Wang et al. proposed a layered and integrated soft sensor model structure, using GMM for sample division , use the random resampling strategy and the partial mutual information criterion to divide and select the variables to achieve the diversity of the enhanced model, and finally use the PLS pruning technology to remove the redundant model. process verified

Although the random resampling strategy can enhance the generalization ability of the model, there is a certain degree of uncertainty in the model structure

Images