Process manufacturing industry irregular sampling dynamic sequence modeling method based on sampling interval perception long and short term memory network

A technology of long short-term memory and sampling interval, applied in biological neural network models, manufacturing computing systems, neural learning methods, etc., which can solve problems such as difficulty for operators to maintain sampling frequency, irregular time length, sampling measurement, etc.

Active Publication Date: 2020-10-27
CENT SOUTH UNIV
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Difficulty for process operators to maintain uniform sampling frequency
Even for some quality variables that can be measured by on-line analyzers, in many process manufacturing industrial processes frequent maintenance in individual process units often results in sampling measurements of irregular lengths of time

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Process manufacturing industry irregular sampling dynamic sequence modeling method based on sampling interval perception long and short term memory network
  • Process manufacturing industry irregular sampling dynamic sequence modeling method based on sampling interval perception long and short term memory network
  • Process manufacturing industry irregular sampling dynamic sequence modeling method based on sampling interval perception long and short term memory network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0130] see Figure 1-5 As shown, a method for modeling irregular sampling dynamic sequences in the process manufacturing industry based on sampling interval-aware long-term short-term memory network includes the following steps:

[0131] Step (1), selecting 43 variables (as shown in Table 1) that have an impact on the C5 content of light naphtha at the initial boiling point of jet fuel from the hydrocracking process as input variables, extracted from September 15, 2016 1,300 samples obtained from offline testing from March 1st to February 9th, 2018.

[0132] Step (2), normalize the data collected in step (1) to obtain a new data set, the conversion function is:

[0133]

[0134] where x min is the minimum value of the data set, x max is the maximum value of the dataset. Standardization of deviation is a linear transformation of the original data, so that the result falls into the [0,1] interval;

[0135] The first 1000 samples are used as the training set to train the ...

Embodiment 2

[0175] The final boiling point of heavy naphtha is predicted as follows:

[0176] Step (1), select 43 variables (as shown in Table 1) that affect the end boiling point of heavy naphtha from the hydrocracking process as input variables, and extract the data from September 15, 2016 to November 2018 871 samples obtained from offline testing on the 30th.

[0177] Step (2), normalize the data collected in step (1) to obtain a new data set, the conversion function is:

[0178]

[0179] where x min is the minimum value of the data set, x max is the maximum value of the dataset. Standardization of deviation is a linear transformation of the original data, so that the result falls into the [0,1] interval;

[0180] The first 632 samples are used as the training set to train the model parameters, and the remaining 239 samples are used as the test set to test the predictive performance of the model. First, the input and output matrices of the training set are obtained:

[0181] ...

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

No PUM Login to view more

Abstract

The invention provides a process manufacturing industry irregular sampling dynamic sequence modeling method based on a sampling interval perception long-term and short-term memory network. The methodspecifically comprises the following steps: firstly, selecting a key process variable which influences the production process and the product quality from the production process as a quality variable,and then continuously and irregularly sampling the input process variable and the quality variable to obtain a dynamic data sequence; preprocessing the sampled original dynamic data sequence; duringmodeling, converting the sampling interval into a proper weight by using a non-additive function, performing calculating by using a full connection layer to obtain a predicted value of a quality variable, and determining training set data and a test data set according to a sequence; training a network, and determining a network structure and hyper-parameters; and realizing real-time online prediction of the quality variable. According to the method, irregular sampling data in the process manufacturing industry can be processed, nonlinear dynamic characteristics in the industry can also be processed, the calculated amount is small, and the applicability and accuracy of the soft measurement model are greatly improved.

Description

technical field [0001] The invention relates to the field of industrial process prediction and control, in particular to a method for modeling irregular sampling dynamic sequences in the process manufacturing industry based on sampling interval sensing long-term and short-term memory networks. Background technique [0002] The process industry includes industries such as petroleum, chemical industry, non-ferrous metals, iron and steel, and building materials, and is an important pillar industry of the national economy. In industrial processes, quality monitoring and control is crucial to process safety, optimization and energy saving; quality monitoring and control largely depends on the real-time online measurement of key performance indicators of the process; due to the harsh measurement environment, online analysis The online measurement of key performance indicators of the process is difficult due to expensive instruments and large delays in offline measurement. As a re...

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
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/08G06Q10/06G06Q50/04
CPCG06N3/049G06N3/084G06Q10/06395G06Q50/04G06N3/044G06N3/045Y02P90/30
Inventor 袁小锋李林王雅琳王凯阳春华桂卫华
Owner CENT SOUTH UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap