The traditional intelligent diagnosis methods of rotating machinery generally require feature extraction of the raw signals in advance. However, it is a very time-consuming and laborious process for extracting the sensitive feature information to improve classification performance. Deep learning method, as a novel machine learning approach, can simultaneously achieve feature extraction and pattern classification. With the characteristics of Deep Belief Network (DBN) and one-dimensional Convolutional Neural Network (1D-CNN) (e.g. learning complex nonlinear, sparse connection and weight sharing), a precise diagnosis method based on the combination of DBN and 1D-CNN is proposed. Firstly, the DBN composed of three pre-trained restricted Boltzmann machines (RBMs) is constructed to achieve feature extraction and dimensionality reduction of the high-dimensional raw data. Secondly, the low-dimensional features extracted by DBN are fed into 1D-CNN for further extracting the abstract features. Finally, Soft-max classifier is employed to identify different operating conditions of rotating machinery. The superiority of the proposed method is validated by comparison with several state-of-the art fault diagnosis methods on two experimental cases. Meanwhile, the proposed method is tested in different background noises and on the imbalanced datasets. The results show that it has higher efficiency and accuracy than the state-of-the art fault diagnosis methods.

Introduction

With the rapid development of science and technology, rotating machinery in modern industry has been moving toward high speed, super precision and high efficiency [1], [2]. After a long-term operating in the complex working environment, the core components of rotating machinery, including gears and bearings, are prone to cause various unperceivable faults. If not detected and managed, these failures may affect the operation of the whole rotating machinery and cause huge economic losses to enterprises [3]–[۵]. Therefore, It’s urgent for us to develop some advanced diagnosis methods, which can accurately and efficiently detect the potential faults of the key components of rotating machinery [6], [7]. At present, there are many methods used in fault diagnosis of rotating machinery, including oil debris analysis, electrical signature analysis, acoustic emission detection, vibration signal analysis, temperature analysis and so on [8]. In contrast with the other approaches, the vibration signal analysis is more common, and the relevant researches are more mature [9], [10]. Additionally, the vibration signals of rotating machinery usually carry more valuable information. A complete fault diagnosis method based on pattern recognition consists of three steps: signal preprocessing [11], feature extraction [12] and pattern classification [13]. Each step has a critical impact on the final recognition accuracies of the model [14].