Fault diagnosis method for secure transmission of industrial data based on transfer learning

A fault diagnosis method for secure transmission of industrial data based on transfer learning, comprising:

• s1:
• collecting raw data from different sources in industrial processes and classifying different types of data into different levels based on importance of the raw data;
• s2:
• constructing a data storage module by defining an ontology model and mapping the raw data from s1 to the ontology model to obtain mapped data;
• s3:
• preprocessing the mapped data in s2 to obtain preprocessed data;
• s4:
• encrypting and transmitting the preprocessed data in s3 to obtain encrypted transmission data;
• s5:
• constructing a transfer learning model and inputting the encrypted transmission data in s4 into the transfer learning model for training to obtain a trained transfer learning model;
• and
• s6:
• using the trained transfer learning model obtained in s5 to achieve fault diagnosis under secure transmission conditions of industrial data;
• wherein the transfer learning model in s5 comprises:
• a feature extraction module, a domain adaptation module, and a diagnostic classification module;
• wherein s5 comprises:
• s51:
• constructing the feature extraction module for feature extraction of the encrypted transmission data, weighting through a Convolutional Block Attention Module (CBAM) hybrid attention mechanism to splice feature maps in a channel dimension, and finally outputting classification results;
• s52:
• constructing the domain adaptation module, wherein cross domain feature distributions at global and local levels are measured by the domain adaptation module through joint maximum mean difference and correlation alignment;
• s53:
• constructing a total loss function for performing reverse updating on the transfer learning model, wherein a joint maximum mean difference loss function, a correlation alignment loss function, and a source domain classification loss function jointly form the total loss function;
• wherein the domain adaptation module comprises:
• the joint maximum mean difference and the correlation alignment, and wherein alignment of the cross domain feature distributions at global and local levels is achieved through the joint maximum mean difference and the correlation alignment;
• wherein in s52, a processing procedure of the domain adaptation module comprises:
• the joint maximum mean difference adjusts global statistical characteristics and local feature relationships of source and target domains through jointly matching edge distributions and conditional distributions;
• wherein an expression is:
• JMMD
• 2
• (
• D
• s
• ,
• D
• t
• )
• =
• 
• E
• x
• s
• ∼
• D
• s
• [
• k
• ⁡
• (
• x
• s
• ,
• ·
• )
• ]
• -
• E
• x
• t
• ∼
• D
• t
• [
• k
• ⁡
• (
• x
• t
• ,
• ·
• )
• ]
• 
• ℋ
• 2
• wherein Ds is a data distribution of a source domain, which represents a distribution of labeled data used during model training;
• Dt is a data distribution of target domain, which represents a distribution of unlabeled data that the transfer learning model needs to adapt to;
• Exs~Ds is an expectation of all samples xs in Ds, that is, a statistical average of source domain data;
• Ext~Dt is an expectation of all samples x, in Dt, that is, a statistical average of target domain data;
• ∥·∥ is a norm;
• k(xs,·) is a kernel function;
• and is a reproducing kernel Hilbert space;
• wherein the correlation alignment is achieved by aligning a covariance matrix of features of the source domain and a covariance matrix of features of the target domain;
• assuming that a data feature of the source domain is XS∈Rns×d and a data feature of the target domain is Xt∈Rnt×d, an interdomain alignment is achieved by minimizing a difference between the covariance matrix of the source domain and the covariance matrix of the target domain, wherein an expression is:
• CORAL
• ⁡
• (
• X
• S
• ,
• X
• t
• )
• =
• 
• Cov
• (
• X
• S
• )
• -
• Cov
• (
• X
• t
• )
• 
• F
• 2
• wherein, Cov(Xs) is the covariance matrix of the source domain;
• Cov(Xt) is the covariance matrix of the target domain;
• and ∥·∥F is a Frobenius norm, used for measuring difference between covariance matrices.