Deep Transfer Network With Multi-Kernel Dynamic Distribution Adaptation for Cross-Machine Fault Diagnosis

Mingzhu Lv,Changzheng Chen,Shixun Liu,Xiaoming Su

doi:10.1109/access.2021.3053075

Abstract

Recently, various deep learning models, which are mainly based on data-driven algorithms, have received more and more attention in the field of intelligent fault diagnosis and prognostics. However, there are two major assumptions accepted by default in the existing studies: 1) The training (source domain) and testing (target domain) data sets obey the same feature distribution; 2) Sufficient labeled data with fault information is available for model training. In real industrial scenarios, especially for different machines, these assumptions are mostly invalid, which makes it a huge challenge to build reliable diagnostic model. Motivated by transfer learning, we present a novel intelligent method named deep transfer network (DTN) with multi-kernel dynamic distribution adaptation (MDDA) to address the problem of cross-machine fault diagnosis. In the proposed approach, the DTN has wide first-layer convolutional kernel and several small convolutional layers, which is utilized to extract transferable features across different machines and suppress high frequency noise. Then, the MDDA method constructs a weighted mixed kernel function to map different transferable features to a unified feature space, and the relative importance of the marginal and conditional distributions are also evaluated dynamically. The proposed method is verified by three transfer learning tasks of bearings, in which the health states of wind turbine bearings in real scenario are identified by using diagnosis knowledge from two different bearings in laboratories. The results show that the proposed method can achieve higher diagnosis accuracy and better transfer performance even under different noisy environment conditions than many other state-of-the-art methods. The presented framework offers a promising approach for cross-machine fault diagnosis.

Highlights

Rolling element bearings are key components of the rotating machinery, whose health states directly affect the performance, stability and service life of the machinery
A new method named multi-kernel dynamic distribution adaptation (MDDA) constructs a weighted mixed kernel function to map the transferable features to a unified high-dimensional feature space, and dynamically evaluates the relative importance of marginal probability distribution (MPD) and conditional probability distribution (CPD), so as to minimize the discrepancy between the two distributions, and obtain a target classifier by the structural risk minimization (SRM) principle
In this paper, a new intelligent fault diagnosis method based on transfer learning named deep transfer network (DTN) with MDDA is presented

Summary

INTRODUCTION

Rolling element bearings are key components of the rotating machinery, whose health states directly affect the performance, stability and service life of the machinery. Different from traditional machine learning methods, the goal of transfer learning is to enhance the performance of the model and reduce the quantity of required sample in the target domain using transferable features or diagnosis knowledge from source domain [9]. Chatterjee and Dethlefs [13] utilized an exponential linear activation function to improve the quality of mapped vibration data, and adopted non-negative constraints to modify the loss function so as to improve the effect of feature-based transfer learning These methods only minimize the distance between cross-domain feature distributions, and do not realize the distribution alignment. On the basis of absorbing and drawing upon informed research, this paper presented a novel intelligent fault diagnosis framework, named deep transfer network (DTN) with multi-kernel dynamic distribution adaptation (MDDA) for cross-machine fault diagnosis.

PROBLEM DESCRIPTION

3) TRAINING PROCEDURE

CASE 1

CASE 2

Findings

CONCLUSION