Precise Fault Diagnosis Research Articles

A fault diagnosis method for rolling bearings based on inter-class repulsive force discriminant transfer learning

For fault diagnosis of rolling bearings, it is generally difficult or even impossible to obtain class labels of new working condition samples under actual variable working conditions, which leads to a low fault diagnosis accuracy. On account of this, we propose a novel unsupervised transfer learning method called inter-class repulsive force discriminant transfer learning (IRFDTL) in this paper. In the proposed IRFDTL, to reduce the classification error in source domain and improve the generalization ability of IRFDTL, a nonnegative extended slack matrix is creatively constructed to transform the strict binary label matrix into an extended slack label matrix. Moreover, the joint distribution discrepancy is introduced to reduce the difference between source and target domains, and the inter-class repulsive force term is innovatively designed to promote the discriminative learning effect by increasing the inter-class distance. Finally, the whole framework of IRFDTL is optimized by the alternating direction multiplier method. By using the labeled samples under historical working conditions, the IRFDTL can perform high-precision class discrimination on the testing samples under new working conditions when there are no class labels of testing samples. The proposed IRFDTL-based fault diagnosis method can achieve precise fault diagnosis of the testing samples under new working conditions, and fault diagnosis instances of rolling bearings verify the effectiveness of the proposed method.

MODELS, DETECTION METHODS, AND CHALLENGES IN DC ARC FAULT: A REVIEW

The power generation of solar photovoltaic (PV) technology is being implemented in every nation worldwide due to its environmentally clean characteristics. Therefore, PV technology is significantly growing in the present applications and usage of PV power systems. Despite the strength of the PV arrays in power systems, the arrays remain susceptible to certain faults. An effective supply requires economic returns, the security of the equipment and humans, precise fault identification, diagnosis, and interruption tools. Meanwhile, the faults in unidentified arc lead to serious fire hazards to commercial, residential, and utility-scale PV systems. To ensure secure and dependable distribution of electricity, the detection of such hazards is crucial in the early phases of the distribution. In this paper, a detailed review of modern approaches for the identification of DC arc faults in PV is presented. In addition, a thorough comparison is performed between various DC arc-fault models, characteristics, and approaches used for the identification of the faults.

LEFE-Net: A Lightweight Efficient Feature Extraction Network With Strong Robustness for Bearing Fault Diagnosis

High precision and fast fault diagnosis is an important guarantee for the safe and reliable operation of machinery. In recent years, due to the strong recognition ability, data-driven fault diagnosis technology based on deep learning has attracted enormous attention. The fault diagnosis module proposed by many scholars has achieved excellent recognition results, but some of them are too complex to deploy in practice, due to the high costs. In this article, an efficient feature extraction method based on the convolutional neural networks (CNN) was proposed, and the high-precision fault diagnosis task was completed used a lightweight network. First, a 1-D CNN (1-D-CNN) is used to extract the multichannel features for the input original signals, to generate the feature maps which are highly consistent with the 2-D CNN (2-D-CNN). Second, a strategy of the lightweight model is proposed based on the principle of dynamic convolution and separable convolution. Third, a spatial attention mechanism (SAM) is used to adjust the weights of the output feature maps. Finally, the model completes the diagnosis task through the learned features. The performance of the model was verified under different operating conditions and noise environments. The experimental results demonstrate that the proposed method has excellent anti-noise ability and domain adaptability.

A Comparison Fault Diagnosis Algorithm for Star Networks

Fault diagnosis for a multiprocessor system is a process of identifying the faulty nodes in the system and is an important issue on the reliability of the system. As to the problem that there are few effective algorithms to diagnose faulty nodes in a given star network system in the literature, this paper proposes a precise fault diagnosis algorithm to identify faulty nodes in a star network system with a given syndrome under the comparison model. Such an algorithm contains three main parts. In the first part, we present an algorithm called Partition-Cycle for partitioning a cycle into sequences based on a given syndrome of the cycle. In the second part, we introduce an algorithm called Digout to diagnose these cycle sequences obtained the first part, which can diagnose each node in the cycle to be faulty or fault-free or unknown. In the third part, we design a diagnosis algorithm called Star-Digout to diagnose faulty nodes in an n-dimensional ( n\geqslant 6) star networks, which is proved to contain a cycle that contains all nodes in the network and is not the same two nodes. Our theoretical analysis shows the time complexity of the diagnosis algorithm is O(n!). Our simulation results show that our algorithm is a precise diagnosis algorithm for a star network system.

An Integrated Deep Learning Method towards Fault Diagnosis of Hydraulic Axial Piston Pump.

A hydraulic axial piston pump is the essential component of a hydraulic transmission system and plays a key role in modern industry. Considering varying working conditions and the implicity of frequent faults, it is difficult to accurately monitor the machinery faults in the actual operating process by using current fault diagnosis methods. Hence, it is urgent and significant to investigate effective and precise fault diagnosis approaches for pumps. Owing to the advantages of intelligent fault diagnosis methods in big data processing, methods based on deep learning have accomplished admirable performance for fault diagnosis of rotating machinery. The prevailing convolutional neural network (CNN) displays desirable automatic learning ability. Therefore, an integrated intelligent fault diagnosis method is proposed based on CNN and continuous wavelet transform (CWT), combining the feature extraction and classification. Firstly, CWT is used to convert the raw vibration signals into time-frequency representations and achieve the extraction of image features. Secondly, a new framework of deep CNN is established via designing the convolutional layers and sub-sampling layers. The learning process and results are visualized by t-distributed stochastic neighbor embedding (t-SNE). The results of the experiment present a higher classification accuracy compared with other models. It is demonstrated that the proposed approach is effective and stable for fault diagnosis of a hydraulic axial piston pump.

Integrated decision‐making method for power transformer fault diagnosis via rough set and DS evidence theories

Precise power transformer fault diagnosis involves incorporating multi-source monitoring information. Uncertain information, missing data, usually occurs in transformer fault cases and diagnosis tasks. To address these challenges, the authors proposed an integrated method of comprehensive transformer fault diagnosis. Diagnostic transformer rules extracted from fault cases form a decision-making table, whereby the main transformer monitoring information and fault types serve as conditional and decision attributes, respectively. Different fault-warning symptoms of the conditional attributes and corresponding decision attributes constitute diagnostic rules. Each obtained symptom in a diagnostic task is evidence supporting different fault types. A modified basic probability assignment (BPA) calculation method is proposed to determine the fault type probability by the obtained symptom. To address contradictory evidence, the symptom significance is introduced to design an improved combination rule incorporating all calculated BPA values to accomplish fault diagnosis. The obtained diagnostic results indicate that more symptoms and a higher symptom significance enable reliable transformer fault diagnosis. The recognition rate of the authors’ method reaches 91.2% with 12–14 symptoms and 94.3% for a 0.9 symptom significance coefficient. It is demonstrated that compared with other combination rules, their method attains a suitable performance (contradiction coefficient K = 0.9 at an 81.3% recognition rate) in realising contradictory information fusion.

Microgrid Fault Detection and Classification: Machine Learning Based Approach, Comparison, and Reviews

Accurate fault classification and detection for the microgrid (MG) becomes a concern among the researchers from the state-of-art of fault diagnosis as it increases the chance to increase the transient response. The MG frequently experiences a number of shunt faults during the distribution of power from the generation end to user premises, which affects the system reliability, damages the load, and increases the fault line restoration cost. Therefore, a noise-immune and precise fault diagnosis model is required to perform the fast recovery of the unhealthy phases. This paper presents a review on the MG fault diagnosis techniques with their limitations and proposes a novel discrete-wavelet transform (DWT) based probabilistic generative model to explore the precise solution for fault diagnosis of MG. The proposed model is made of multiple layers with a restricted Boltzmann machine (RBM), which enables the model to make the probability reconstruction over its inputs. The individual RBM layer is trained with an unsupervised learning approach where an artificial neural network (ANN) algorithm tunes the model for minimizing the error between the true and predicted class. The effectiveness of the proposed model is studied by varying the input signal and sampling frequencies. A level of considered noise is added with the sample data to test the robustness of the studied model. Results prove that the proposed fault detection and classification model has the ability to perform the precise diagnosis of MG faults. A comparative study among the proposed, kernel extreme learning machine (KELM), multi KELM, and support vector machine (SVM) approaches is studied to confirm the robust superior performance of the proposed model.

Study on uncertainty reasoning of running reliability testing face to CNC machine

It is necessary to test the reliability of the operation of the machine itself for meeting the high-speed, ultra-precision, flexible, and other modern manufacturing and processing needs. In practical applications, the obtained reliability detection information presents a variety of uncertainties such as randomness, imprecision, incompleteness and so on, due to the complexity of the machining environment, the limitations of the sensor itself and the imperfections of the information acquisition technology. The uncertainty reasoning method is proposed to test reliability and accuracy precision. Based on the multi-sensor information characteristic, the rule of the credibility sets is established to achieve the operational reliability analysis. It can realise the uncertainty reasoning for the modal and reliability rules. Examples of analysis show that during the high-end CNC machine tools operational reliability analysis, it has a practical effect to study the uncertainty reasoning method of the modal and reliability propositional rule and the analysis basis can be provided for the precision degradation and fault diagnosis.

The Design of Intelligent Fault Detection System for an Automatic Loader

The fault characteristic of automatic loading system is analysed, the fuzzy-immunity mixed fault diagnosis method is put forward, and the programme of fuzzy-immunity fault diagnosis is given, the math model of fault diagnosis is established, the fault diagnosis is resolved in the lack of prior knowledge. The precision fault diagnosis is realized for the automatic loading system.

Wind Turbine Diagnosis under Variable Speed Conditions Using a Single Sensor Based on the Synchrosqueezing Transform Method.

The gearbox is one of the key components in wind turbines. Gearbox fault signals are usually nonstationary and highly contaminated with noise. The presence of amplitude-modulated and frequency-modulated (AM-FM) characteristics compound the difficulty of precise fault diagnosis of wind turbines, therefore, it is crucial to develop an effective fault diagnosis method for such equipment. This paper presents an improved diagnosis method for wind turbines via the combination of synchrosqueezing transform and local mean decomposition. Compared to the conventional time-frequency analysis techniques, the improved method which is performed in non-real-time can effectively reduce the noise pollution of the signals and preserve the signal characteristics, and hence is suitable for the analysis of nonstationary signals with high noise. This method is further validated by simulated signals and practical vibration data measured from a 1.5 MW wind turbine. The results confirm that the proposed method can simultaneously control the noise and increase the accuracy of time-frequency representation.

Finite-sensor fault-diagnosis simulation study of gas turbine engine using information entropy and deep belief networks

Precise fault diagnosis is an important part of prognostics and health management. It can avoid accidents, extend the service life of the machine, and also reduce maintenance costs. For gas turbine engine fault diagnosis, we cannot install too many sensors in the engine because the operating environment of the engine is harsh and the sensors will not work in high temperature, at high rotation speed, or under high pressure. Thus, there is not enough sensory data from the working engine to diagnose potential failures using existing approaches. In this paper, we consider the problem of engine fault diagnosis using finite sensory data under complicated circumstances, and propose deep belief networks based on information entropy, IE-DBNs, for engine fault diagnosis. We first introduce several information entropies and propose joint complexity entropy based on single signal entropy. Second, the deep belief networks (DBNs) is analyzed and a logistic regression layer is added to the output of the DBNs. Then, information entropy is used in fault diagnosis and as the input for the DBNs. Comparison between the proposed IE-DBNs method and state-of-the-art machine learning approaches shows that the IE-DBNs method achieves higher accuracy.

Improved Precise Fault Diagnosis Algorithm for Hypercube-Like Systems Based on the Comparison Diagnosis Model

Multiprocessor systems are being increasingly adopted and the system reliability is an important perspective for multiprocessor systems. The fault diagnosis has become crucial for achieving high reliability in multiprocessor systems. The precise fault diagnosis diagnoses all processors correctly. In the comparison-based model, it allows a processor to perform diagnosis by contrasting the responses from a pair of neighboring processors through sending the identical assignment. On the basis of comparison-based model, Sengupta and Dahbura (“On self-diagnosable multiprocessor systems: diagnosis by the comparison approach,” IEEE Transaction on Computers, vol. 41, no. 11, pp. 1386–1396, 1992) put forward the MM* model, any processor c diagnoses two processors c1 and c2 if c has direct communication links to them. Sengupta and Dahbura also designed an O(N5)-time precise fault diagnosis algorithm to diagnose faulty processors for general topologies by using the MM* model, where N is the cardinality of processor set in multiprocessor systems. Lately, Ye and Hsieh (“A scalable comparison-based diagnosis algorithm for hypercube-like net-works,” IEEE Transaction on Reliability, vol. 62, no. 4, pp. 789–799, 2013) devised an precise fault diagnosis algorithm to diagnose all faulty processors for hypercube-like networks by using the MM* model with O(N(log2N)2) time complexity. On the basis of Hamiltonian cycle properties, we improve the aforementioned results by presenting an O(N)-time precise fault diagnosis algorithm to diagnose all faulty processors for hypercube-like networks by using the MM* model.

Fault feature extraction and classification based on WPT and SVD: Application to element bearings with artificially created faults under variable conditions

Achieving a precise fault diagnosis for rolling bearings under variable conditions is a problematic challenge. In order to enhance the classification and achieves a higher precision for diagnosing rolling bearing degradation, a hybrid method is proposed. The method combines wavelet packet transform, singular value decomposition and support vector machine. The first step of the method is the decomposition of the signal using wavelet packet transform and then instantaneous amplitudes and energy are computed for each component. The Second step is to apply the singular value decomposition to the matrix constructed by the instantaneous amplitudes and energy in order to reduce the matrix dimension and obtaining the fault feature unaffected by the operating condition. The features extracted by singular value decomposition are then used as an input to the support vector machine in order to recognize the fault mode of rolling bearings. The method is applied to a bearing with faults created using electro-discharge machining under laboratory conditions. Test results show that the proposed methodology is effective to classify rolling bearing faults with high accuracy.

Study on a New Fault Diagnosis Method Based on Combining Intelligent Technologies

In allusion to more indeterminate information and higher speed request characteristic in fault diagnosis system, according to the intelligence complementary strategy, a new fault diagnosis(SWPSO-BPN) model based on combining improved particle swarm optimization (PSO) algorithm and Back-propagation(BP) neural network is proposed in this paper. In the SWPSO-BPN method, an improved PSO (SWPSO) algorithm is proposed to optimize the parameters of BP neural network in order to overcome the shortcomings of slow learning speed and being easy to fall into local minimum, and obtain the optimal values of parameter combination in the BP neural network(SWPSOBPN) model. Then proposed SWPSO-BPN model is applied to diagnose the fault in order to obtain a new fault diagnosis (SWPSO-BPN-FD) method. Finally, the proposed SWPSO-BPN-FD method is used to test the data from bearing data center of CWRU. The experimental results show that the proposed SWPSO-BPN-FD method can accurately and effectively realize high precision fault diagnosis of rolling bearing. And this method takes on strong robustness and generalization ability.

A Deadlock-Free and Connectivity-Guaranteed Methodology for Achieving Fault-Tolerance in On-Chip Networks

To improve the reliability of on-chip network based systems, we design a deadlock-free routing technique that is more resilient to component failures and guarantees a higher degree of node connectivity. The routing methodology consists of three key steps. First, we determine the maximal connected subgraph of the faulty network by checking whether the defective components happen to be the cut vertices and bridges of the network topology. A precise fault diagnosis mechanism is used to identify partial defective routers. Second, we construct an acyclic channel dependency graph that breaks all cycles and preserves connectivity of the maximal connected subgraph . This is done through the cycle-breaking and connectivity guaranteed (CBCG) algorithm. Finally, we introduce a fault-tolerant adaptive routing scheme that can be used with or without virtual channels for network congestion avoidance and high-throughput routing. The simulation results show both the effectiveness and robustness of the proposed approach. For an $8 \times 8$ 2D-Mesh with 40 percent of link damage, full connectivity and deadlock freedom are still archived without disabling any faultless router in 98.18 percent of the simulations. In a 2D-Torus, the simulation percentage is even higher (99.93 percent). The hardware overhead for supporting the introduced features is minimal. An on-line implementation of cbcg using TSMC 65nm library has only 0.966 and 1.139 percent area overhead for the $8\times8$ and $16 \times 16$ 2D-Meshes.

Multi-State Adaptive BIT False Alarm Reduction Under Degradation Process

Built-in tests (BITs) are widely used in mechanical systems to detect and diagnose a fault, whereas the BIT false alarms bring much trouble for precise fault diagnosis and logistics/maintenance arrangement. The false alarm phenomenon is related to the degradation over time, and the false alarm evolution process can be typically divided into three stages. This paper proposes a condition-based multistage false alarm detection and reduction method for mechanical systems. The stages are clarified according to the degradation level and the false alarm severity. The dividing boundaries of the stages are optimized using soft margin one-versus-rest support vector machine (SVM) classifiers. The associated intermediate stage is the intense period of false alarms, and the dynamic Bayesian network inference model is developed to satisfy the requirements of accurate false alarm diagnosis. To achieve the goal of false alarm suppression, the top-level BIT outputs are updated with the original BIT alarms and the identified probable states. Finally, the proposed approach is demonstrated in the application study of a milling machine and the well-round experimental results are analyzed.

Precision fault diagnosis procedure for a structural system having a defect employing Hidden Markov Models

Condition-based maintenance (CBM) is one of the most effective maintenance methods since it can minimize operation shutdowns for maintenance. In CBM, a fault diagnosis is conducted using vibration signals caused by various fault types. To analyze signal variations automatically and efficiently, pattern recognition methods are usually employed. Typical pattern recognition methods employed for CBM include the Artificial Neural Network (ANN) and the Hidden Markov Model (HMM). Among them, HMMs were employed in this paper to identify the size and location of a crack in a structural system. In CBM employing HMMs, feature vector extraction is the most important step for reliable diagnosis. In this paper, Fast Fourier Transform (FFT) was employed for feature vector extraction. Even though it is relatively easy to identify the fault type, it is difficult to identify the size and location of a defect since signal variation due to defect size or location variation is extremely small. In this paper, a diagnosis procedure to identify the size and location of a defect is proposed. The effectiveness of the proposed method is validated using a numerical analysis model of a rotating blade having a crack.

State Evaluation of Diesel Engine Using Genetic Algorithm for the Feature Selection and Optimize

As for diesel engine fault diagnosis, it is more difficult to select the appropriate feature. In the paper, using the main parts state such as cylinder, piston and piston ring to diagnosis the fault of diesel engine. Eigenvalue is vital to the precision and efficiency fault diagnosis, the normal method to find the fault diagnosis eigenvalue is simulation experiment. Since the number of experiment cannot be too much, bootstrap is introduced in the paper, then using Genetic Algorithm, an optimum parameter combination has been received.

Exact Modeling and Simulation of Saturated Induction Motors with Broken Rotor Bars Fault using Winding Function Approach

Winding function method (WFM) provides a detailed and rather simple analytical modeling and simulation technique for analyzing performance of faulty squirrel-cage induction motors (SCIMs). Such analysis is mainly applicable for designing on-line fault diagnosis techniques. In this paper, WFM is extended to include variable degree of magnetic saturation by applying an appropriate air gap function and novel techniques for estimating required saturation factor and angular position of the air gap flux density. The resultant saturable winding function method (SWFM) is properly applied for analyzing SCIMs with broken rotor bars and for identifying the saturation effect on the related fault indexes. Comparing simulation results to the corresponding experimental results show higher accuracy of the SWFM. This means that more precise fault diagnosis techniques can be designed by using the proposed saturable modeling technique.

Rolling Bearing Fault Feature Extraction Based on SVD-EEMD

The novel method that singular value decomposition (SVD) is combined with ensemble empirical mode decomposition (EEMD) is proposed because of the mode mixing in empirical mode decomposition (EMD). The first step of this method is to reduce the random noise in fault signal by the SVD, and then does EEMD to restrain the mode mixing effectively. Finally, the intrinsic mode function (IMF) is done for envelope demodulation and as a result, the fault feature is extracted successfully. The implementation process was analyzed by simulation signal and this method has been successfully applied to in inner race and outer race of rolling bearing fault diagnosis. The results show that this method can extract the fault information of rolling bearing effectively and realize the precise fault diagnosis.