Neural Network Based Fault Diagnosis Research Articles

Multiscale holospectrum convolutional neural network-based fault diagnosis of rolling bearings with variable operating conditions

In recent years, the convolutional neural network (CNN) has been widely used in the field of intelligent fault diagnosis. However, each convolutional layer of CNN cannot take the overall and local information into account, and the feature extraction ability of CNN with fewer layers is weak. These circumstances lead to poor performance of CNN in practical fault diagnosis with variable operating conditions. To solve these problems, this paper proposes a multiscale holospectrum CNN (MH-CNN) based on the methods of two-dimensional multiscale feature fusion and decision-level feature fusion. First, the continuous wavelet transform is used to map the time-domain signal to the time-frequency plane to fully reflect the complex information contained in the signal. Then the two-dimensional multiscale feature fusion is introduced to extract features at different scales, which can take both overall and local information into account. Finally the decision-level feature fusion is introduced to fuse the features from signal in X, Y directions in the decision-level of CNN, which serves to enhance the features. By combining these methods, the proposed MH-CNN can extract more distinguishable features with a shallow structure, which can ensure the classification capability while avoiding the overfitting problem caused by overly complex networks. The effectiveness of the MH-CNN is verified using complicated data sets consisting of 16 rolling bearings with four different health conditions, two speeds and three loads. Results show that the proposed MH-CNN achieves a correct rate of 99.8% for rolling bearing fault diagnosis under variable operating conditions, which is much higher than other comparative methods.

Artificial neural network–based fault diagnosis for induction motors under similar, interpolated and extrapolated operating conditions

The diagnosis of mechanical and electrical faults of induction motors (IMs) has been performed using artificial neural networks (ANN) for similar, interpolated and extrapolated operating speeds. The current and vibration signals of faulty and healthy IMs measured from a Machinery Fault Simulator are used in this work. In total, ten different IM fault conditions have been considered: four mechanical faults (bearing fault, unbalanced rotor, misaligned rotor, and bowed rotor), five electrical faults (broken rotor bar, phase unbalanced fault with two severity levels, and stator winding fault with two severity levels), and one healthy motor condition. An ANN model is developed in which raw time domain data of faulty IMs are used and the fault diagnosis is then performed for the motor’s various operating conditions. Initially, diagnosis is performed to predict and classify the motor faults, for the same operating conditions for which we trained ANN. The diagnosis is then extended for interpolated and extrapolated speeds in order to accomplish the diagnosis when data are not available at all the required operating speeds. From the results, it is found that the present ANN-based diagnosis is effective in the same speed case for various operating conditions (seven speeds as well as three loads). In addition, the diagnosis is found to be satisfactory for all interpolated and extrapolated speed cases. It is also observed that the present IM fault diagnosis is better in the interpolation speed cases than the extrapolation speed cases.

An explainable one-dimensional convolutional neural networks based fault diagnosis method for building heating, ventilation and air conditioning systems

Due to the frequently changed outdoor weather conditions and indoor requirements, heating, ventilation and air conditioning (HVAC) experiences faulty operations inevitably throughout its lifespan. Therefore, it is important to monitor and diagnose HVAC fault operations. Recently, deep learning methods have attracted more attentions for their guarantee of better diagnosis performance under various system configurations and operating conditions. However, these methods are black-box models which though highly accurate for fault diagnosis but are extremely hard to explain. To overcome the disadvantage of poor interpretability of deep learning black-box models, this study therefore proposes a novel explainable deep learning based fault diagnosis method that is suitable for HVACs. To maintain HVAC operational information and variable locations of all chiller input data samples, proposed method is established with three characteristics: 1) the pooling layer is excluded, 2) the size of convolution filter kernel is set as 1, and 3) use softsign as an activation function. Considering the resulting impacts of HVAC faults on system operating variables, a new Absolute Gradient-weighted Class Activation Mapping (Grad-Absolute-CAM) method is proposed to visualize the fault diagnosis criteria and make the model explainable by providing the fault-discriminative information. The proposed method is validated using fault experimental dataset of a typical building HVAC system (i.e., chiller) from the ASHRAE research project 1043 (RP-1043). The fault diagnosis accuracy is over 98.5% for seven chiller faults. Results indicates that it is capable of interpreting the model work mechanism by activation feature maps and explaining the fault diagnosis criteria by Grad-Absolute-CAM.

Optimized multi layer perceptron artificial neural network based fault diagnosis of induction motor using vibration signals

Optimized multi layer perceptron artificial neural network based fault diagnosis of induction motor using vibration signals

Deep residual network for enhanced fault diagnosis of rotating machinery

Deep residual network (DRN) is a recently-developed powerful algorithm in the deep learning filed. This paper introduces the superiority of DRN into the fault diagnosis of rotating machinery for simplifying traditional diagnosing process as well as enhancing predicting performance. DRN can not only extract features automatically from raw or processed signals but also benefit from its especially deep architecture to continually improve representation capacity without worrying gradient divergence issues. The functions of DRN result from the unique structure design called residual building block, which will be described clearly with the network overall architecture in this paper. Additionally, the comparison of the DRN with other machine learning-based and neural network-based fault diagnosis methods are presented.

Neural Network-Based Fault Diagnosis of Joints in High Voltage Electrical Lines

Neural Network-Based Fault Diagnosis of Joints in High Voltage Electrical Lines

Small‐batch‐size convolutional neural network based fault diagnosis system for nuclear energy production safety with big‐data environment

Small‐batch‐size convolutional neural network based fault diagnosis system for nuclear energy production safety with big‐data environment

Review on Deep Learning Based Fault Diagnosis

The massive high-dimensional measurements accumulated by distributed control systems bring great computational and modeling complexity to the traditional fault diagnosis algorithms, which fail to take advantage of the higher-order for online estimation. In view of its powerful ability of representation learning, deep learning based fault diagnosis is extensively studied, both in academia and in industry, making intelligent process control more automated and effective. In this paper, deep learning based fault diagnosis is reviewed and summarized as four parts, i.e., stacked auto-encoder based fault diagnosis, deep belief network based fault diagnosis, convolutional neural network based fault diagnosis, and recurrent neural network based fault diagnosis. Furthermore, some necessity and potential trends, integrated innovation, + knowledge and information fusion, are discussed from the view of data preprocessing, network design and decision.

A robust intelligent fault diagnosis method for rolling bearings based on Deep convolutional neural network and Domain Adaptation

Abstract At present, it is difficult for the deep neural network based fault diagnosis to obtain satisfactory fault classification accuracy when the labeled fault data is insufficient under different working conditions. We proposed a robust intelligent fault diagnosis method for rolling bearings based on Deep convolutional neural network and Domain Adaptation. We first created pair-wised samples using a large amount of labeled source data and a small amount of unlabeled target data. Based on the 1D convolutional neural network, a Deep Convolution Siamese Adaptation network (DCSAN) is designed to map the input space of the two domains data into a common feature space. By defining contrastive loss over multiple adaptation layers, a multi-layer adaptation method is proposed. We designed a balance factor to weigh the importance of different adaptation layers, which can achieve a better adaptation performance than one-layer adaptation. The experimental results show that the method can effectively improve the accuracy of fault diagnosis when there is only a small amount of vibration data from target domain under variable working conditions.

Multi-stream convolutional neural network-based fault diagnosis for variable frequency drives in sustainable manufacturing systems

Fault detection and diagnosis of induction motors in variable frequency drive (VFD) applications is essential for minimizing unexpected downtime, material waste and equipment damage, ultimately contributing to sustainable manufacturing. This paper presents a multi-stream convolutional neural network (MS-CNN) for automatic feature extraction from and fusion of motor vibration and stator current at various line frequencies. The MS-CNN has demonstrated superior performance over conventional machine learning methods. To understand the rationale for MS-CNN to diagnose motor defects, the relevance of input features for fault classification by a trained MS-CNN are investigated through Layer-wise Relevance Propagation (LRP) of its predictions.

A novel automata and neural network based fault diagnosis system for PLC controlled manufacturing systems

The Fault and Anomaly Detection and Isolation (FADI) in Programmable Logic Controller (PLC) controlled systems is an important and challenging problem. In this paper, we present an automated tool, called the Manufacturing Process Failure Diagnosis Tool (MPFDT) that can detect and isolate the faults and anomalies in the PLC controlled manufacturing systems effectively. MPFDT utilizes two independent knowledge-based process behaviour models of the manufacturing system to satisfy the FADI purpose. The fundamental idea is to detect the inconsistencies between the modelled and the observed manufacturing process behaviour. The first model is a Deterministic Finite-state Automaton (DFA) based model of the PLC control process that is used to determine whether the observed state transition behaviour of the PLC control process is consistent with the modelled state transition behaviour or not. The second model is basically a set of Artificial Neural Network (ANN) based one-class classifiers that are used to identify whether any significant difference exist between the observed and the reference electrical power consumption profile of the manufacturing system or not. The experimental results show that the FADI accuracy rate of the proposed tool is very high (more than 98%).

A Lighted Deep Convolutional Neural Network Based Fault Diagnosis of Rotating Machinery.

To improve the fault diagnosis performance for rotating machinery, an efficient, noise-resistant end-to-end deep learning (DL) algorithm is proposed based on the advantages of the wavelet packet transform in vibration signal processing (the capability to extract multiscale information and more spectral distribution features) and deep convolutional neural networks (good classification performance, data-driven design and high transfer-learning ability). First, a vibration signal is subjected to pyramid wavelet packet decomposition, and each sub-band coefficient is used as the input for each channel of a deep convolutional network (DCN). Then, based on the lightweight modeling requirements and techniques, a new DCN structure is designed for the fault diagnosis. The proposed algorithm is compared with the support vector machine algorithm and the published DL algorithms based on a bearing dataset produced by Case Western Reserve University. The experimental results show that the proposed algorithm is superior to the existing algorithms in terms of accuracy, memory space, computational complexity, noise resistance, and transfer performance, producing good results.

Artificial Neural Network Based Fault Diagnosis of a Pulley-Belt Rotating System

Rotating machines are widely used in various industrial fields. Hence, an unexpected stoppage due to, for example, bad operating conditions or manufacturing error, has safety implications along with economic considerations. In this research, a fault detection system for a pulley-belt rotating system is developed and then different faults simulated in a test rig are investigated. Vibration signal monitoring is utilized since it represents a reliable approach for fault recognition in rotating machinery. Time-domain signal analysis technique is applied to extract some indicative features, such as root mean square, kurtosis and skewness. An artificial neural network (ANN) model is developed to detect the simulated faults. However, in addition to the machine healthy condition five fault types, such as unbalance in the driving pulley, wear in the belt and pulleys misalignment, have been simulated in the test rig. Two MEMS accelerometers (ADXL335), interfaced to Arduino MEGA 2560 as a data acquisition device, are used for vibration amplitude measurement. LabVIEW, which is a graphical programming software, is utilized to develop a signal capturing, analysis and feature extraction system. The result showed the effectiveness of the developed system in detection of different fault types in the pulley-belt system.

A deep neural network based fault diagnosis method for centrifugal chillers

Various types of faults occur in building energy systems throughout their life-cycles. Some faults grow gradually causing system energy penalty and performance degradation. Hence, it is crucial to implement an efficient fault diagnosis strategy and maintain optimal operations for systems. Recently, data-driven methods have got increasing interests due to the model flexibility and data availability. The fast development of data science has provided advanced data analytics to tackle data classification problems in a more convenient and efficient way. This paper attempts to investigate the potential of a promising data analysis technique, i.e., deep neural network, in classifying and diagnosing faults in a building energy system, i.e., centrifugal chiller plant. This study exploits the deep neural network based method in both supervised and unsupervised manners, and compares the fault diagnosis accuracy. Centrifugal chiller experimental data from the ASHRAE Research Project 1043(RP-1043) are used to validate the proposed method. Results show that the method can correctly diagnoses the fault data for seven typical chiller faults.

Convolution Neural Network Based Fault Diagnosis of Induction Motor

Convolution Neural Network Based Fault Diagnosis of Induction Motor

Optimized neural network-based fault diagnosis strategy for VRF system in heating mode using data mining

This paper presents the optimized back propagation neural network (BPNN) method for fault diagnosis of the variable refrigerant flow air conditioning (VRF) system in the heating mode. A feature variable set optimization approach of diagnosis models is proposed based on data mining method. First, the correlation analysis method is used to eliminate redundant variables. Then, the association rule mining method is used to optimize the feature set (FS) selection. Five FSs (FS1-FS5) are obtained by optimized feature variable selection. FS1 is the original set. FS2 is the set obtained by correlation analysis and FS3-FS5 are the sets obtained by the association rule mining. The fault diagnosis models with different FSs are evaluated using four fault experiments which include outdoor unit heat exchanger air-side fouling, four-way reversing valve fault, refrigerant undercharge and refrigerant overcharge faults. The results show that the correlation analysis method can effectively eliminate redundant variables and the association rule mining method is feasible to optimize the FSs for fault diagnosis. The BPNN-FS5 model shows the best fault diagnosis performance of the VRF system in the heating mode, whose fault diagnosis correct rate has increased from 88.71% to 96.40% and hit rates of four faults are higher than 90%.

Composite Fault Diagnosis in Wireless Sensor Networks Using Neural Networks

Wireless sensor networks (WSNs) are spatially distributed devices to support various applications. The undesirable behavior of the sensor node affects the computational efficiency and quality of service. Fault detection, identification, and isolation in WSNs will increase assurance of quality, reliability, and safety. In this paper, a novel neural network based fault diagnosis algorithm is proposed for WSNs to handle the composite fault environment. Composite fault includes hard, soft, intermittent, and transient faults. The proposed fault diagnosis protocol is based on gradient descent and evolutionary approach. It detects, diagnose, and isolate the faulty nodes in the network. The proposed protocol works in four phases such as clustering phase, communication phase, fault detection and classification phase, and isolation phase. Simulation results show that the proposed protocol performs better than the existing protocols in terms of detection accuracy, false alarm rate, false positive rate, and detection latency.

Thermal analysis MLP neural network based fault diagnosis on worm gears

The importance of fault diagnoses, in any kind of machinery, can’t be over stated. Any undetected small fault in machinery will most probably rise with time and will cause machinery to shut down thus resulting in both mechanical and more importantly economical loss for the industry. In recent years, researches have been done for the faults diagnosis through the analysis of their vibration and sound signatures. The extraction of those characteristic signatures is a complicated process because complexities in modern day machineries can results in many vibration and sound generating sources. This paper presents a condition based fault diagnoses technique to detect the condition of gear. An experimental setup, consisting of a worm gear driven by an electric motor, was setup to conduct tests under different working conditions. The vibration and sound signature signals of worm gear were examined for normal and faulty conditions under different speeds and oil levels. The collected data was then used for feature extraction, by using Fast Fourier Transform to filter background noise signals and to collect only the signature of the gearbox vibration and sound signals. An MLP (Multilayer Perceptron) Artificial Neural Network Model has been developed to classify the signature signals. A thermal camera is also used to observe the heating patterns for all those working conditions. With the help of MLP Artificial Neural Network it is possible to predict the speed and oil level of the gearbox and hence a possible fault diagnoses is also feasible.

Validation of Neural Network-based Fault Diagnosis for Multi-stack Fuel Cell Systems: Stack Voltage Deviation Detection

This paper presents (i) an algorithm for the detection of unexpected stack voltage deviations in an Solid Oxide Fuel Cells (SOFC)-based power system with multiple stacks and (ii) its validation in a simulated online environment. The algorithm is based on recurrent neural networks (RNNs) and is validated by using operating data from the Wärtsilä WFC20 multi-stack SOFC system. The voltage deviation detection is based on statistical testing. Instead of a hardware implementation in the actual power plant, the algorithm is validated in a simulated online environment that provides data I/O communication based on the OPC (i.e. Object Linking and Embedding (OLE) for Process Control) protocol, which is also the technology utilized in the real hardware environment. The validation tests show that the RNN-based algorithm effectively detects unwanted stack voltage deviations and also that it is online-viable.

Fault diagnosis of rotating machinery based on kernel density estimation and Kullback-Leibler divergence

Based on kernel density estimation (KDE) and Kullback-Leibler divergence (KLID), a new data-driven fault diagnosis method is proposed from a statistical perspective. The ensemble empirical mode decomposition (EEMD) together with the Hilbert transform is employed to extract 95 time- and frequency-domain features from raw and processed signals. The distance-based evaluation approach is used to select a subset of fault-sensitive features by removing the irrelevant features. By utilizing the KDE, the statistical distribution of selected features can be readily estimated without assuming any parametric family of distributions; whereas the KLID is able to quantify the discrepancy between two probability distributions of a selected feature before and after adding a testing sample. An integrated Kullback- Leibler divergence, which aggregates the KLID of all the selected features, is introduced to discriminate various fault modes/damage levels. The effectiveness of the proposed method is demonstrated via the case studies of fault diagnosis for bevel gears and rolling element bearings, respectively. The observations from the case studies show that the proposed method outperforms the support vector machine (SVM)-based and neural network-based fault diagnosis methods in terms of classification accuracy. Additionally, the influences of the number of selected features and the training sample size on the classification performance are examined by a set of comparative studies.