Multi-sampling rate soft sensor method based on dynamic latent variable model

Abstract

The invention discloses a multi-sampling rate soft sensing method based on a dynamic hidden variable model. By taking a large number of process variables with different sampling rates in chemical processes and a small number of key quality variables as the modeling samples, dynamic latent variables which can contain the characteristics of multi-rate data are extracted while fully considering the autocorrelation and cross-correlation of data, and the estimation of model parameters is realized by an expectation maximization algorithm and a Kalman filter algorithm. Based on this model, the corresponding soft-sensing method is established to solve the estimation problem of multi-rate dynamic key quality variables. The method not only realizes the multi- sampling rate information processing, but also can make full use of the data information. Moreover, the dynamic characteristics of the data can be fully considered by a Kalman filter, and the dynamic latent variables can be accurately estimated, so that the estimation and description of the key quality variables which are difficult to be directly measured can be better realized for a few of the dynamic latent variables after dimension reduction, and the soft sensing accuracy and application range can be improved.

Description

technical field [0001] The invention designs a control method, in particular relates to a multi-sampling-rate soft-sensing method based on a dynamic hidden variable model. Background technique [0002] With the development of modern industry, process safety and product quality are widely valued. With the widespread application of distributed control systems (DCS) in the industrial field, a large number of process variables can be collected and stored by various high-sampling rate sensors, while key quality variables related to production safety and product quality need to be stored in a low-sampling rate method. Collecting and obtaining through assays leads to the multi-sampling rate characteristics of data and the difficulty of obtaining important variable data, which is a challenge for the management of actual industrial engineering. At the same time, with the continuous advancement of multivariate statistical analysis-based process monitoring (MSPM) and soft-sensing tech...

AI Technical Summary

Problems solved by technology

Traditional static Principal Component Analysis (PCA) and Least Square Estimation (PLS) models can effectively extract the cross-correlation of variables, but are not effective when the measurement is highly time-dependent

Dynamic PCA (DPCA)-based techniques can extract eigenvalue decomposition on autocorrelation augmented matrices in measurements to more effectively handle data dynamics, but it cannot fully utilize multi-sampled data

Both canonical variable analysis (CVA) and PLS can simulate the relationship between two data sets and realize the soft measurement of data, but both cannot effectively deal with the dynamics of data, and cannot deal with the multi-sampling characteristics of the two data sets themselves. Case

However, the method based on multi-sampling probability principal component analysis can fully utilize the multi-sampling rate data information, and use the expectation maximization (EM) algorithm to effectively estimate the model parameters, but the processing effect on dynamic data is not good.

Images