A Quality Prediction and Monitoring Method Based on Multivariate Mutual Information Optimization
A quality prediction and mutual information technology, applied in data processing applications, complex mathematical operations, instruments, etc., can solve problems such as local optimality of intelligent optimization algorithms
Active Publication Date: 2021-09-10
浙江天信咨询监理有限公司
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Although the use of intelligent optimization algorithms such as genetic algorithms can select input variables related to quality from the overall level, it is a well-known problem that intelligent optimization algorithms are prone to fall into local optimum
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 Construction
[0039] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0040] Such as figure 1 As shown, the present invention discloses a quality prediction and monitoring method based on multivariate mutual information optimization, and the specific implementation of the method is as follows.
[0041] First, the offline modeling phase includes steps (1) to (11) as shown below.
[0042] Step (1): From the historical database of the production process object, find out the data corresponding to the index that can reflect the product quality to form the output matrix Y∈R n×k , the sampled data corresponding to the output Y constitutes the input matrix X∈R n×m .
[0043] Step (2): Calculate the mean μ of each column vector in the output matrix Y 1 , μ 2 ,…, μ k and standard deviation δ 1 ,δ 2 ,…,δ k After that, according to the formula Standardize the row vectors in Y to get the normalized output matrix ...
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 a quality prediction and monitoring method based on multi-variable mutual information optimization, aiming to solve how to optimally select process measurement variables related to quality indicators from the overall level based on multi-variable mutual information, and establish corresponding quality indicators based on this Indicator forecasting and monitoring models. By enumerating all possible combinations of input variables, the method of the invention can absolutely ensure that the optimal input variables related to output are selected, and can avoid the problem of variable optimization falling into local optimum. In addition, the method of the present invention utilizes the optimized input variables to establish a soft-sensing model, which can eliminate the interference effect of measurement data irrelevant to the quality index. Because the method of the invention not only implements the soft measurement of the quality index, but also implements the real-time monitoring of the quality index according to the soft measurement value, and distinguishes the faults into those related to quality and not related to quality. Therefore, the method of the invention can better solve the quality-related soft measurement and monitoring problems.
Description
technical field [0001] The invention relates to a quality prediction and monitoring method, in particular to a quality prediction and monitoring method based on multivariate mutual information optimization. Background technique [0002] In the design of the entire integrated automation system, product quality is the primary consideration, and the system for real-time monitoring of product quality occupies a very important position. With the rapid development of science and technology, although measuring instruments such as temperature, pressure, flow, etc. high. Taking the quality index of concentration as an example, the price of an online analyzer for measuring concentration components is more than ten times that of ordinary instruments for measuring temperature and pressure, and manual maintenance is required in the later period. If the online analyzer is not used to obtain quality data in real time due to cost reasons, product quality data can be obtained through offli...
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): G06Q10/06G06F17/16
CPCG06F17/16G06Q10/06395
Inventor 宋励嘉童楚东俞海珍
Owner 浙江天信咨询监理有限公司