Chemical production on-line fault detection and diagnosis technique based on physical-large data hybrid model

A fault detection and hybrid model technology, applied in chemical data mining, electrical digital data processing, special data processing applications, etc., can solve problems such as the increase of fault diagnosis error rate and the mistrust of production technicians in fault detection.

Active Publication Date: 2016-08-24
陆新建
View PDF3 Cites 30 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But for chemical production, many unit operations, especially physical processes (such as heat exchange, compression, rectification, etc.), have a definite mass and energy balance relationship, and there are often physical isolation between the front and rear operation units of the process (such as being separated by the middle The storage tank isolates two separation units), there is no physical correlation between physical parameters, and the big data analysis technology is directly used, because the noise effect between the measurement data, or is affected by the utility (such as power steam temperature, fuel Influenced by atmospheric pressure composition), these physically unrelated data often have a high degree of correlation in the model established by big data, which makes big data analysts mistakenly believe that there is a correlation. This correlation that violates the laws of physics makes the fault diagnosis The increase in misjudgment rate has led to distrust of production technicians in fault detection, which has brought certain resistance to the promotion of Industry 4.0 and the construction of smart factories

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
  • Chemical production on-line fault detection and diagnosis technique based on physical-large data hybrid model
  • Chemical production on-line fault detection and diagnosis technique based on physical-large data hybrid model
  • Chemical production on-line fault detection and diagnosis technique based on physical-large data hybrid model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] Embodiment 1: Detection and diagnosis of the stability failure of the system.

[0046] In order to test the effectiveness of the new method, the normal data and abnormal data sets are marked in the data set for testing, in which the normal data contains 61225 data points and 66 data points with stability failures (at 55370-55436 positions),

[0047] The first step is to use the real-time database of the device to collect data to form a training sample set for modeling: X R 61291×112 . Among them, the list is shown in Attached Table 1.

[0048] The second step targets production units with a defined physical model and extracts key performance indicators based on the physical model.

[0049] The feed heating furnace in the production process is taken as an example. For details, see figure 2 , in furnace unit Q, there are about 28 process variables, Xq R 61291×28 Some process variables are uncontrollable. For example, heat dissipation is usually related to the ins...

Embodiment 2

[0076] Embodiment 2: detection and diagnosis of the performance fault of this system:

[0077] Next, in conjunction with this specific process, the implementation steps of the present invention are described in detail:

[0078] The first step. Same as the first step in Example 1.

[0079] The second step; the same as the second step in Example 1.

[0080] The third step is to conduct a double-layer regression analysis on the performance approval standard to obtain a monitoring model for performance failure.

[0081] After extracting the key indicators of the physical model and reducing the dimensionality, Xnew containing relevant physical information is obtained and normalized (same example 1), and the relationship between Xnew and plant-level key performance indicators is regressed to monitor whether there is a performance failure in the process.

[0082] to attach figure 1Taking the production process in China as an example, the key performance indicators at the factory l...

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 discloses a chemical production on-line fault detection and diagnosis technique based on a physical-large data hybrid model. The chemical production on-line fault detection and diagnosis technique is characterized by comprising the following steps: after selecting a target operation unit, scanning all historical data of the unit, and after parameters are verified, establishing an accident knowledge base and a parameter model; in later on-line detection process, directly loading on-line data into the parameter model, scanning so as to obtain fault data in the on-line data, making an alarm, and comparing the fault data with data in the accident knowledge base, thereby obtaining fault reasons. The method disclosed by the invention is applied to confirmed single chemical unit operation, a reliable physical model can be established, a chemometrical method is used in the whole production process, a big-data processing technique is introduced to process real-time data of years, operators focus on faults of performance indexes within a controllable variable range, and the influence of uncontrollable production process variable to fault detection is eliminated.

Description

technical field [0001] The invention relates to the field of safety detection and control in the chemical production process, in particular to an online fault detection and diagnosis technology for chemical production based on a physical-big data hybrid model. Background technique [0002] The chemical production process is extremely complex, accompanied by many unexplored physical and chemical reactions. For the continuous process industry, maintaining a stable working condition can not only reduce accidents and secondary disasters, but also stabilize product quality and operate at optimal cost under optimized working conditions to obtain maximum economic benefits. Therefore, the faults referred to here include not only abnormal conditions affecting device safety (collectively referred to as safety faults), but also deviations in product quality and optimal operating conditions (collectively referred to as performance faults). [0003] For complex chemical production proce...

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): G06F17/50G06F19/00G06K9/62
CPCG16C20/70G06F30/20G06F18/214
Inventor 陆新建
Owner 陆新建
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