Alarm source seeking method based on data driving

A data-driven, root-based technology, applied in the direction of electrical digital data processing, special data processing applications, instruments, etc., can solve the problems of high requirements for the number of modeling data, reduce major accidents, improve economic benefits, and apply strong effect

Inactive Publication Date: 2017-06-20
QUANZHOU INST OF EQUIP MFG
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The shortcoming of this type of method is that it has high requirements for the amount of modeling data, and the characteristics of modern industrial systems that generate massive data just make up for this shortcoming

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
  • Alarm source seeking method based on data driving
  • Alarm source seeking method based on data driving
  • Alarm source seeking method based on data driving

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] combine figure 1 To illustrate this specific implementation mode, a data-driven method for finding the root cause of an alarm disclosed in Embodiment 1 of the present invention is carried out in the following steps:

[0051] Step 1: Detect the working data of the industrial system and obtain the observed variables, and store the d observed variables in the data matrix X. This embodiment uses the augmented Fuller test method to check the time stationarity of the data, and performs Preprocessing, preprocessing includes the use of filtering and other methods to process data noise; the working data includes parameters that reflect the operation of the system, such as temperature, pressure, water level, etc.;

[0052] Step 2: Initialize the model parameters, and use the Cao criterion to optimize the model parameters; the model refers to the model that establishes the causal relationship of variables, and the model parameters are some setting parameters needed to establish th...

Embodiment 2

[0105] Embodiment 2: This embodiment is different from the specific embodiment 1 in that: Step 2 adopts Ragwitz criterion to optimize parameters.

specific Embodiment

[0106] Specific embodiment: the method for finding the root cause of an alarm based on data-driven in this specific embodiment is used to simulate the causality of variables in a flue gas desulfurization process (flue gas desulfurization, FGD) of an oil company, and the specific steps are as follows;

[0107] Step 1, taking the FDG process as an example, select the liquid levels of the reaction tank, tank 1 and tank 2, and the flow rates of pumps 2 and 3 as variables, denoted as y 1 、y 2 、y 3 the y 4 、y 5 , collect 3544 sets of data, the data has time stationarity, and preprocess the data;

[0108] Step 2, model parameters are initialized, and the parameters are optimized using the Cao criterion;

[0109] Step 3, calculate the TE value and NTE value between variables, see Table 1;

[0110] Table 1

[0111]

[0112] Select 0.02 as the threshold, and the information flow path based on the standard transition entropy is as follows: Figure 5 shown;

[0113] Step 4, cal...

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 an alarm source seeking method based on data driving. According to the method, through data collection, data time stability checking, data pre-processing and parameter optimization setting, a mathematical model of relationships among variables is established, then an NTE value and an NDTE value among the variables are calculated, and therefore whether or not a direct causal relationship exists among system variables is judged. The method does not rely on a system physical model and priori knowledge, a causal relationship can be obtained by only relying on the process measurement variables, and the method is high in applicability and can be widely applied to industrial fields such as chemical industries, textile and metallurgy. A source can be sought in an initial stage of alarm giving, the fault isolation and elimination can be carried out conveniently and in real time, the accident occurrence is reduced or even avoided, the safety and reliability of system running are improved, and at the same time the environment pollution can be reduced.

Description

technical field [0001] The invention belongs to the technical field of safety monitoring, in particular to a data-driven method for finding the root cause of an alarm. Background technique [0002] Due to the continuous improvement of the requirements for the safety and reliability of industrial systems, online and real-time monitoring of the system operation process has become an indispensable key link in modern industrial systems. Considering that it is difficult to obtain accurate mathematical models and prior knowledge of the system, and industrial systems generate a large amount of historical operating data, data-driven process monitoring has become the mainstream technology of modern industrial safety monitoring. Sending out an alarm after a fault occurs can help the staff judge the operation of the system in time, but this method cannot determine the cause of the alarm. Finding the root cause of the alarm can clarify the cause of the alarm when the alarm occurs, so i...

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): G06F17/50
CPCG06F30/20
Inventor 陈豪张景欣王耀宗张丹蔡品隆
Owner QUANZHOU INST OF EQUIP MFG
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