An Adaptive Soft Sensor Method Based on Semi-Supervised Incremental Gaussian Mixture Regression
A Gaussian mixture, semi-supervised technology, applied in adaptive control, instruments, control/regulation systems, etc., can solve the problem of inaccurate model parameter learning, improve prediction accuracy, reduce prediction error, and alleviate over-fitting effects.
Active Publication Date: 2022-04-29
ZHEJIANG UNIV
View PDF9 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Assuming that there is a linear relationship between the process variable and the quality variable, the probability density function, regression coefficient and mixing coefficient of each component are learned through the Expectation Maximization (EM) algorithm, and the Bayesian Information Criterion (BIC) is used for model selection to effectively solve the industrial The problem of inaccurate model parameter learning caused by the scarcity of labeled samples in the process
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0102] The performance of the semi-supervised incremental Gaussian mixture regression model is illustrated below with an example of a reformer in the production process of a specific hydrogen production unit in the ammonia synthesis process. The main raw material NH3 of the hydrogen production unit in the ammonia synthesis process is usually the main raw material in the urea synthesis process. According to the design of the process flow, the primary reformer is the main place for the conversion reaction. The process flow chart is as follows figure 2 shown. According to the reaction mechanism, the reaction temperature is the key factor to ensure the production of hydrogen in the first-stage reformer. In order to stabilize the temperature at a certain level, it is necessary to monitor the combustion state in real time. It is necessary to control the oxygen content at the top of the furnace within the set range. one of the effective means. In an actual industrial process, the c...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses an adaptive soft sensor method based on semi-supervised incremental Gaussian mixture regression. The method first uses an incremental Gaussian mixture regression model, and selects a group of process variables that are highly correlated with key quality variables and are easy to measure As input to the model, predictive estimation of quality variables that are difficult to measure in real-time in time-varying industrial processes is performed. In order to solve the impact of the scarcity of labeled samples in the industrial process on the prediction accuracy of the model, the incremental Gaussian mixture regression model is extended to a semi-supervised incremental Gaussian mixture regression model. The invention can not only effectively face the nonlinear, non-Gaussian and time-varying characteristics in the actual industrial process, but also effectively solve the problem of inaccurate model parameter learning caused by the scarcity of labeled samples in the industrial process, and alleviate the problem to a certain extent. Over-fitting of the model is prevented, and the efficiency of model update is improved, and the purpose of adaptive soft sensor for key variables is achieved.
Description
technical field [0001] The invention belongs to the field of industrial process prediction and control, in particular to an adaptive soft sensor method based on semi-supervised incremental Gaussian mixture regression. Background technique [0002] In the actual industrial production process, there are often more or less key process variables that cannot be detected online. In order to solve this problem, a variable that is easier to detect in the collection process is constructed according to an optimal standard. With these variables as input and key process variables as output mathematical model, the online estimation of key process variables is realized. This is the soft sensor modeling commonly used in industrial processes. [0003] The development of statistical process soft-sensing modeling methods has an extremely significant demand for large-scale industrial data. Among them, the Gaussian mixture regression model can well solve the nonlinear and non-Gaussian character...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G05B13/04
CPCG05B13/042
Inventor 宋执环李德阳
Owner ZHEJIANG UNIV