Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Anomaly detection method of industrial control network signal based on deep learning structure

An industrial control network and deep learning technology, applied in the field of outlier detection in industrial control network data, can solve problems such as difficult detection of a small number of outliers, and achieve the effect of improved ability

Active Publication Date: 2021-05-04
HARBIN INST OF TECH
View PDF8 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to solve the problem that in the existing method, it is difficult to detect a small number of abnormal values ​​and it is necessary to artificially define the difference between normal data and abnormal values.

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
  • Anomaly detection method of industrial control network signal based on deep learning structure
  • Anomaly detection method of industrial control network signal based on deep learning structure
  • Anomaly detection method of industrial control network signal based on deep learning structure

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0024] Specific implementation mode one: refer to Figure 1 to Figure 4 Specifically illustrate this embodiment, the industrial control network signal anomaly detection method based on the deep learning structure described in this embodiment, the method includes the following steps:

[0025] Step 1. Select part of the data from the industrial control network data to mark as a training sample, perform data normalization and standardization operation on the training sample, obtain the normalized calibrated data, and use the data enhancement algorithm to normalize the normalized calibrated data. The calibration data adds some false sample values ​​to form the detected data;

[0026] Step 2, input the normal data and the detected data into an autoencoder compression network for training, respectively obtain the spliced ​​data of the normal data and the spliced ​​data of the detected data;

[0027] Step 3, input the two spliced ​​data into the comparison network and calculate the ...

specific Embodiment approach 2

[0044] Specific embodiment 2: This embodiment is a further description of the industrial control network signal anomaly detection algorithm based on the deep learning structure described in the specific embodiment 1. In this embodiment, in step 1, data normalization is performed on the training samples The formula for the normalization operation is:

[0045]

[0046] In the formula, is the mathematical expectation of the training sample data, σ x is the standard deviation of the training sample data, x k is the normalized calibrated data, x m is the calibrated data before normalization;

[0047] In step 1, the formula for adding some fake samples to the normalized calibrated data using the data enhancement algorithm is:

[0048]

[0049]

[0050] In the formula, λ is a value in the range of (0,1), (x i ,y i ),(x j ,y j ) are two samples randomly obtained from the normalized calibrated data; is the value and label of the generated "fake samples".

specific Embodiment approach 3

[0051] Specific embodiment three: This embodiment is a further description of the industrial control network signal anomaly detection algorithm based on the deep learning structure described in specific embodiment one. In this embodiment, in step three, the two spliced ​​data Input to the comparison network and calculate the distance between the normal data and the detected data through the calculation of the deep neural network is as follows:

[0052] Using the parameter θ c neural network to learn the difference L(f n ,f a ; θ c ), the distance f(x) between the normal data and the detected data is obtained as:

[0053] f(x)=L([x n ,z n ],[x a ,z a ]; θ c ) Formula 3,

[0054] where f n represented by normal data X n and the compressed value Z for normal data n Splicing data of normal data obtained by splicing and combining, f a Indicates that by the new detected data X a and the detected data compression value Z a The spliced ​​data of the detected data obtain...

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 an abnormal detection method for industrial control network signals based on a deep learning structure, and relates to the technical field of abnormal value detection in industrial control network data. The present invention aims to solve the problem that it is difficult to detect a small amount of abnormal values ​​because it needs to be artificially defined for distinguishing normal data and abnormal values ​​in the existing methods. Select part of the data from the industrial control network data as a training sample, perform data normalization and standardization operations on the training sample, obtain the normalized calibrated data, and use the data enhancement algorithm to add some false positives to the normalized calibrated data The sample values ​​form the detected data; the normal data and the detected data are input into an autoencoder compression network for training, and the trained data are obtained respectively; the data are input into the comparison network and calculated by the deep neural network to obtain The distance between the normal data and the detected data, using a classifier to determine the abnormal value in the detected data according to the distance. It is used for signal anomaly detection.

Description

technical field [0001] The invention relates to an industrial control network signal abnormality detection method based on a deep learning structure, and belongs to the technical field of abnormal value detection in industrial control network data. Background technique [0002] Industrial Control System (Industrial Control System, ICS) refers to an automatic control system composed of computers and industrial process control components, which consists of controllers, sensors, transmitters, actuators, and input / output interfaces. These components are connected through industrial communication lines according to a certain communication protocol to form an industrial manufacturing or processing system with automatic control capabilities. [0003] The current industrial control system usually involves the following types of networks during specific deployment: enterprise office network (enterprise network or office network), process control and monitoring network (monitoring net...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06F2218/12G06F18/214
Inventor 曲海成秦济韬陈浩
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com