An Intelligent Fault Diagnosis Method Using Unsupervised Feature Learning Towards Mechanical Big Data

Lei, Yaguo; Jia, Feng; Lin, Jing; Xing, Saibo; Ding, Steven X.

doi:10.1109/tie.2016.2519325

Cited by 1,033 publications

(512 citation statements)

References 48 publications

Supporting

Mentioning

511

Contrasting

Unclassified

Order By: Relevance

“…It tries to learn good features that satisfy the three principles consisting of population sparsity, lifetime sparsity, and high dispersal [14,27]. Population sparsity means that each sample should be represented by only several active features, namely most of the features extracted from each sample should be zero.…”

Section: Sparse Filteringmentioning

confidence: 99%

“…Recently, artificial intelligence and solutions to extracting features from big data are widely paid attention to. A new tendency to damage manage based on measured data streams typed big data and data-driven models using Artificial Neural Networks (ANN), Fuzzy Logic (FL) or ANFIS [7,8,9,10,11,12,13,14] has been taken form with a promising future. For this aim, in order to improve the ability to process, analyze and extract valuable information from the initial vibration data sets, wavelet and SSA are also often used [10,11].…”

Section: Introductionmentioning

confidence: 99%

“…In spite of owning some advantages, the three methods as mentioned do not appropriately satisfy online applications due to high calculating cost. Recently, in order to make fault diagnosis less dependent on prior knowledge and diagnostic expertise when processing mechanical big data, an intelligent bearing fault diagnosis methodwas proposed in [14]. The method established a two-stage unsupervised learning process to extract features and estimate system.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Method of Bearing Fault Diagnosis Using Singular Spectrum Analysis, Sparse Filtering and Anfis

Nguyen¹,

Nguyen

2018

JCC

View full text Add to dashboard Cite

Abstract. Bearing is an important machine detail participating in almost all mechanical systems.Estimating online its operating condition to exploit actively the systems, therefore, is one of the most urgent requirements. This paper presents an online bearing damage identifying method named ASBDIM based on ANFIS (Adaptive Neuro-Fuzzy Inference System), Singular Spectrum Analysis (SSA) and sparse filtering. This is an online estimating process operated via two phases, offline and online one. In the offline period, by using SSA and sparse filtering, a database signed Off DaB is built whose inputs are features extracted from the measured data stream typed big data, while its outputs are values encoding the surveyed bearing damage statuses. The ANFIS is then employed to identify the dynamic response of the mechanical system corresponding to the bearing damage statuses reflected by the Off DaB. In the online period, first, at each estimating time, another database called On DaB is established using the way similar to the one used for building the Off DaB. The On DaB participates as inputs of the ANFIS to generate its outputs which are then compared with the corresponding encoded outputs to specify bearing real status at this time. Survey results based on different data sources showed the effectiveness of the proposed method.Keywords. Identifying bearing damage, AI for estimating damage, ANFIS based damage identification, SSA for identifying damage.

show abstract

Section: Sparse Filteringmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Method of Bearing Fault Diagnosis Using Singular Spectrum Analysis, Sparse Filtering and Anfis

Nguyen¹,

Nguyen

2018

JCC

View full text Add to dashboard Cite

show abstract

“…It is necessary to mine data in the submarine optical fiber network fault diagnosis database and then calculate its attributes to realize big data attribute analysis [10][11][12].By using the tree structure to show the result of data mining, the method is simple and intuitive [13][14], and therefore it is suitable for this paper. The specific process is shown in Figure 1.…”

Section: A Collection and Analysis Of Big Data Attribute Selection Mmentioning

confidence: 99%

Research on Big Data Attribute Selection Method in Submarine Optical Fiber Network Fault Diagnosis Database

Chen

2017

Polish Maritime Research

View full text Add to dashboard Cite

At present, in the fault diagnosis database of submarine optical fiber network, the attribute selection of large data is completed by detecting the attributes of the data, the accuracy of large data attribute selection cannot be guaranteed. In this paper, a large data attribute selection method based on support vector machines (SVM) for fault diagnosis database of submarine optical fiber network is proposed. Mining large data in the database of optical fiber network fault diagnosis, and calculate its attribute weight, attribute classification is completed according to attribute weight

show abstract

“…For example, a two-stage learning method including sparse filtering and neural network was proposed to form an intelligent fault diagnosis method to learn features from raw signals [11]. The feed-forward neural network using Levenberg-Marquardt algorithm showed a new way to detect and diagnose induction machine faults [12], where the results were not affected by the load condition and the fault types.…”

Section: Introductionmentioning

confidence: 99%

A Deep Learning Approach for Fault Diagnosis of Induction Motors in Manufacturing

Shao

Sun

Yan

et al. 2017

Chin. J. Mech. Eng.

134

View full text Add to dashboard Cite

Extracting features from original signals is a key procedure for traditional fault diagnosis of induction motors, as it directly influences the performance of fault recognition. However, high quality features need expert knowledge and human intervention. In this paper, a deep learning approach based on deep belief networks (DBN) is developed to learn features from frequency distribution of vibration signals with the purpose of characterizing working status of induction motors. It combines feature extraction procedure with classification task together to achieve automated and intelligent fault diagnosis. The DBN model is built by stacking multiple-units of restricted Boltzmann machine (RBM), and is trained using layer-bylayer pre-training algorithm. Compared with traditional diagnostic approaches where feature extraction is needed, the presented approach has the ability of learning hierarchical representations, which are suitable for fault classification, directly from frequency distribution of the measurement data. The structure of the DBN model is investigated as the scale and depth of the DBN architecture directly affect its classification performance. Experimental study conducted on a machine fault simulator verifies the effectiveness of the deep learning approach for fault diagnosis of induction motors. This research proposes an intelligent diagnosis method for induction motor which utilizes deep learning model to automatically learn features from sensor data and realize working status recognition.

show abstract

An Intelligent Fault Diagnosis Method Using Unsupervised Feature Learning Towards Mechanical Big Data

Cited by 1,033 publications

References 48 publications

A Method of Bearing Fault Diagnosis Using Singular Spectrum Analysis, Sparse Filtering and Anfis

A Method of Bearing Fault Diagnosis Using Singular Spectrum Analysis, Sparse Filtering and Anfis

Research on Big Data Attribute Selection Method in Submarine Optical Fiber Network Fault Diagnosis Database

A Deep Learning Approach for Fault Diagnosis of Induction Motors in Manufacturing

Contact Info

Product

Resources

About