Research on Rotating Machinery Fault Diagnosis Based on Improved Multi-target Domain Adversarial Network

Aiming at the problems of low efficiency,poor anti-noise and robustness of transfer learning model in intelligent fault diagnosis of rotating machinery,a new method of intelligent fault diagnosis of rotating machinery based on single source and multi-target domain adversarial network model(WDMACN)and Gram Angle Product field(GAPF)was proposed.Firstly,the original one-dimensional vibration signal is preprocessed using GAPF to generate the image data including all time series.Secondly,the residual network is used to extract data features,and the features of the target domain without labels are pseudo-labeled,and the transferable features among the feature extractors are shared through the depth parameter,and the feature extractors of the multi-target domain are updated anatomically to generate the features that the discriminator cannot distinguish.The modelt through adversarial domain adaptation,thus achieving fault classification.Finally,a large number of validations were carried out on the bearing data set of Case Western Reserve University(CWRU)and the gear data.The results show that the proposed method can greatly improve the diagnostic efficiency of the model,and has good noise resistance and generalization.

Compound Fault Diagnosis for Rotating Machinery:State-of-the-Art,Challenges,and Opportunities

Compound fault,as a primary failure leading to unexpected downtime of rotating machinery,dramatically increases the difficulty in fault diagnosis.To deal with the difficulty encountered in implementing compound fault diagnosis(CFD),researchers and engineers from industry and academia have made numerous significant breakthroughs in recent years.Admittedly,many systematic surveys focused on fault diagnosis have been conducted by reputable researchers.Nevertheless,previous review articles paid more attention to fault diagnosis with several single or independent faults,resulting in that there is still lacking a comprehensive survey on CFD.Therefore,to fulfill the above requirements,it is necessary to provide an in-depth overview of fault diagnosis methods or algorithms for compound faults of rotating machinery and uncover potential challenges or opportunities that would guide and inspire readers to devote their efforts to promoting fault diagnosis technology more effective and practical.Specifically,the backgrounds,including the related definitions and a new taxonomy of CFD methods,are detailed according to the way of implementing compound fault recognition.Then,the stateof-the-art applications of CFD are overviewed based on relevant publications in the past decades.Finally,the challenges and opportunities associated with implementing CFD are concluded and followed by a conclusion for ending this survey.We believe that this review article can provide a systematic guideline of CFD from different aspects for potential readers and seasoned researchers.

FAULT DIAGNOSIS OF ROTATING MACHINERY USING KNOWLEDGE-BASED FUZZY NEURAL NETWORK

A novel knowledge-based fuzzy neural network （KBFNN） for fault diagnosis is presented. Crude rules were extracted and the corresponding dependent factors and antecedent coverage factors were calculated firstly from the diagnostic sample based on rough sets theory. Then the number of rules was used to construct partially the structure of a fuzzy neural network and those factors were implemented as initial weights, with fuzzy output parameters being optimized by genetic algorithm. Such fuzzy neural network was called KBFNN. This KBFNN was utilized to identify typical faults of rotating machinery. Diagnostic results show that it has those merits of shorter training time and higher right diagnostic level compared to general fuzzy neural networks.

2D-HIDDEN MARKOV MODEL FEATURE EXTRACTION STRATEGY OF ROTATING MACHINERY FAULT DIAGNOSIS

A new feature extraction method based on 2D-hidden Markov model（HMM） is proposed. Meanwhile the time index and frequency index are introduced to represent the new features. The new feature extraction strategy is tested by the experimental data that collected from Bently rotor experiment system. The results show that this methodology is very effective to extract the feature of vibration signals in the rotor speed-up course and can be extended to other non-stationary signal analysis fields in the future.

On Fault Diagnosis of Rotating Machinery Using Wavelet Time-division Scale Level Moment

Based on an in-depth study of wavelet gray moment, we proposed a concept of a time-division scale level moment and gave the specific definition; ulteriorly, we discussed the factors which affected the fault diagnosis ability of a time-division scale level moment. The analysis results in the caculation of six typical fault signals show that the time-division scale level moment can be used to display the detailed information of a wavelet gray level image, extract the signal＇s characteristics effectively, and distinguish the vibration fault. Compared to the method of a wave gray moment vector, the method mentioned in this paper can provide higher calculation speed and higher capacity of fault identification, so it is more suitable for online fault diagnosis for rotating machinery.

A Method of Rotating Machinery Fault Diagnosis Based on the Close Degree of Information Entropy

This paper presents a method of rotating machinery fault diagnosis based on the close degree of information entropy. In the view of the information entropy, we introduce four information entropy features of the rotating machinery, which describe the vibration condition of the machinery. The four features are, respectively, denominated as singular spectrum entropy, power spectrum entropy, wavelet space state feature entropy and wavelet power spectrum entropy. The value scopes of the four information entropy features of the rotating machinery in some typical fault conditions are gained by experiments, which can be acted as the standard features of fault diagnosis. According to the principle of the shorter distance between the more similar models, the decision-making method based on the close degree of information entropy is put forward to deal with the recognition of fault patterns. We demonstrate the effectiveness of this approach in an instance involving the fault pattern recognition of some rotating machinery.

Study on Fault Diagnosis of Rotating Machinery with Hybrid Neural Networks

ＳｔｕｄｙｏｎＦａｕｌｔＤｉａｇｎｏｓｉｓｏｆＲｏｔａｔｉｎｇＭａｃｈｉｎｅｒｙｗｉｔｈＨｙｂｒｉｄＮｅｕｒａｌＮｅｔｗｏｒｋｓＺａｎｇＣｈａｏｐｉｎｇ（臧朝平）ＧａｏＷｅｉ（高伟）（ＮＥＲＣＴＶ，ＳｏｕｔｈｅａｓｔＵｎｉｖｅｒｓｉｔｙ，Ｎａｎｊｉｎ...

A Modified Feedforward Neural Network Model for Fault Diagnosis of Rotating Machinery

ＡＭｏｄｉｆｉｅｄＦｅｅｄｆｏｒｗａｒｄＮｅｕｒａｌＮｅｔｗｏｒｋＭｏｄｅｌｆｏｒＦａｕｌｔＤｉａｇｎｏｓｉｓｏｆＲｏｔａｔｉｎｇＭａｃｈｉｎｅｒｙＺａｎｇＣｈａｏｐｉｎｇ（臧朝平）ＧａｏＷｅｉ（高）（ＮＥＲＣＴＶ，ＳｏｕｔｈｅａｓｔＵｎｉｖｅｒｓ...

An Approach to Fault Diagnosis of Rotating Machinery Using the Second-Order Statistical Features of Thermal Images and Simplified Fuzzy ARTMAP

Thermal image, or thermogram, becomes a new type of signal for machine condition monitoring and fault diagnosis due to the capability to display real-time temperature distribution and possibility to indicate the machine’s operating condition through its temperature. In this paper, an investigation of using the second-order statistical features of thermogram in association with minimum redundancy maximum relevance (mRMR) feature selection and simplified fuzzy ARTMAP (SFAM) classification is conducted for rotating machinery fault diagnosis. The thermograms of different machine conditions are firstly preprocessed for improving the image contrast, removing noise, and cropping to obtain the regions of interest (ROIs). Then, an enhanced algorithm based on bi-dimensional empirical mode decomposition is implemented to further increase the quality of ROIs before the second-order statistical features are extracted from their gray-level co-occurrence matrix (GLCM). The highly relevant features to the machine condition are selected from the total feature set by mRMR and are fed into SFAM to accomplish the fault diagnosis. In order to verify this investigation, the thermograms acquired from different conditions of a fault simulator including normal, misalignment, faulty bearing, and mass unbalance are used. This investigation also provides a comparative study of SFAM and other traditional methods such as back-propagation and probabilistic neural networks. The results show that the second-order statistical features used in this framework can provide a plausible accuracy in fault diagnosis of rotating machinery.

Application of Kernel GDA to Performance Monitoring and Fault Diagnosis for Rotating Machinery

Faults in rotating machine are difficult to detect and identify,especially when the system is complex and nonlinear.In order to solve this problem,a novel performance monitoring and fault diagnosis method based on kernel generalized discriminant analysis(kernel GDA,KGDA)was proposed.Through KGDA,the data were mapped from the original space to the high-dimensional feature space.Then the statistic distance between normal data and test data was constructed to detect whether a fault was occurring.If a fault had occurred,similar analysis was used to identify the type of faults.The effectiveness of the proposed method was evaluated by simulation results of vibration signal fault dataset in the rotating machinery,which was scalable to different rotating machinery.

FAULT DIAGNOSIS EXPERT SYSTEM FOR ROTATING MACHINERY BASED ON A FUZZY PROBABILITY LOGIC INFERENCE MODEL

A new theory- the fuzzy probability logic theory is presented , This theory incorpo- rates the genterally-used fuzzy logic and the traditionally-used probability logic theory in attempt to emulate the rational fault diagnosis under uncertainty. According to the theory , an inference model , named as FSL , is thus designed to be devoted to the building of a fault diagnosis expert system for rotating machinery (ROSLES) . The system is put into operation on a vibration simula- tor stand for 300 MW turbine generator set ( 1 : 1 0) and satisfactory results are gained.

A Novel Method Based on UNET for Bearing Fault Diagnosis

Reliability of rotating machines is highly dependent on the smooth rolling of bearings.Thus,it is very essential for reliable operation of rotating machines to monitor the working condition of bearings using suitable fault diagnosis and condition monitoring approach.In the recent past,Deep Learning(DL)has become applicable in condition monitoring of rotating machines owing to its performance.This paper proposes a novel bearing fault diagnosis method based on the processing and analysis of the vibration images.The proposed method is the UNET model that is a recent development in DL models.The model is applied to the 2D vibration images obtained by transforming normalized amplitudes of the time-series vibration data samples into the corresponding vibration images.The UNET model performs pixel-level feature learning using the vibration images owing to its unique architecture.The results demonstrate that the model can perform dense predictions without any loss of label information,generally caused by the sliding window labelling method.The comparative analysis with other DL models confirmed the superiority of the UNET model which has achieved maximum accuracy of 98.91%and F1-Score of 99%.

Comparative Study of Combined Fault Diagnosis Schemes Based on Convolutional Neural Network

In this paper, comparative combined fault diagnosis schemes are studied including vibration analysis, acoustic signal analysis and thermal image analysis based on the Convolutional Neural Network (CNN). The advantage of the CNN structure is that it does not need manual feature extraction or selection, which requires prior knowledge of specific machinery dynamics. The vibration and acoustic signals were transformed into spectrograms, which are effective for the diagnostic analysis by using CNN. Comparatively, the thermal images were directly analyzed using CNN. The effectiveness of the CNN-based diagnosis methods was investigated through the analysis of different experimental data, i.e., vibration, acoustic signals and thermal images, which were collected from a test rig where different types of faults are induced on the roller bearing and shaft. The results show that the thermal image analysis and acoustic signal analysis could achieve relatively higher accuracy rate compared to vibration analysis. Moreover, the advantage is easy-deployment because of the non-contact way during signal acquisition. With the CNN-based fault diagnosis method for the three different signals collected, the accuracy of different signal predictions for combined faults can be compared, and the effective method can be applied to fault diagnosis of other industrial rotating machinery.

Rotating machinery fault diagnosis based on convolutional neural network and infrared thermal imaging

Rotating machinery is widely applied in industrial applications.Fault diagnosis of rotating machinery is vital in manufacturing system,which can prevent catastrophic failure and reduce financial losses.Recently,Deep Learning(DL)-based fault diagnosis method becomes a hot topic.Convolutional Neural Network(CNN)is an effective DL method to extract the features of raw data automatically.This paper develops a fault diagnosis method using CNN for InfRared Thermal(IRT)image.First,IRT technique is utilized to capture the IRT images of rotating machinery.Second,the CNN is applied to extract fault features from the IRT images.In the end,the obtained features are fed into the Softmax Regression(SR)classifier for fault pattern identification.The effectiveness of the proposed method is validated using two different experimental data.Results show that the proposed method has a superior performance in identification various faults on rotor and bearings comparing with other deep learning models and traditional vibration-based method.

Imbalanced fault diagnosis of rotating machinery using autoencoder-based SuperGraph feature learning

Existing fault diagnosis methods usually assume that there are balanced training data for every machine health state.However,the collection of fault signals is very difficult and expensive,resulting in the problem of imbalanced training dataset.It will degrade the performance of fault diagnosis methods significantly.To address this problem,an imbalanced fault diagnosis of rotating machinery using autoencoder-based SuperGraph feature learning is proposed in this paper.Unsupervised autoencoder is firstly used to compress every monitoring signal into a low-dimensional vector as the node attribute in the SuperGraph.And the edge connections in the graph depend on the relationship between signals.On the basis,graph convolution is performed on the constructed SuperGraph to achieve imbalanced training dataset fault diagnosis for rotating machinery.Comprehensive experiments are conducted on a benchmarking publicized dataset and a practical experimental platform,and the results show that the proposed method can effectively achieve rotating machinery fault diagnosis towards imbalanced training dataset through graph feature learning.

基于HMFDE和t-SNE的旋转机械故障诊断方法

针对旋转机械的故障特征提取较难,以及单一类型的特征无法全面反映故障特性的问题,提出了一种基于混合多尺度波动散布熵(HMFDE)、t分布-随机邻域嵌入(t-SNE)和郊狼优化算法(COA)优化极限学习机(ELM)的旋转机械故障诊断方法。首先,采用特征加权提出了混合多尺度波动散布熵方法,并将其用于提取旋转机械振动信号的故障特征;随后,采用t-SNE方法对混合故障特征进行了特征降维,挑选出了最能够反映故障特性的特征子集,构建了敏感特征样本;最后,采用郊狼优化算法对极限学习机的输入权重和隐含层阈值进行了优化,完成了旋转机械的故障识别和分类;以齿轮箱和滚动轴承故障数据集为对象,对基于HMFDE、t-SNE和COA-ELM的故障诊断方法进行了实验,验证了方法的有效性。研究结果表明:采用HMFDE-t-SNE-CAO-ELM故障诊断方法可以取得100%的故障识别准确率,该方法能够有效地诊断旋转机械的不同故障类型和损伤;相较于基于单一类型特征的故障诊断方法,其准确率分别可以提高0.68%、22.42%、29.18%(齿轮箱)和1.43%、8.23%、23.67%(滚动轴承),虽然牺牲了一定的计算效率,但准确率得到了明显的提高;相较于其他常规故障分类器,COA-ELM的故障识别准确率具有明显的优势。

工业大数据环境下旋转机械智能故障诊断技术研究

针对旋转机械故障诊断精度较低的问题,提出利用工业大数据对旋转机械的故障进行智能化诊断.研究利用工业大数据对旋转机械的状态进行评估,并且采用深度学习对机械的健康程度进行诊断、监测,对工业大数据环境下的旋转机械诊断的发展趋势进行分析,促使机械故障诊断正式进入大数据时代,实现故障诊断的智能评估.

无量纲与SVM的石化机组旋转机械故障诊断方法

针对石化机组旋转机械故障信息存在非线性、重叠性等特点,提出一种无量纲与支持向量机(Support Vector Machine,SVM)的石化机组旋转机械故障诊断方法。首先对采集的振动信号进行分析并将其无量纲化;接着通过特征选择选取高价值与敏感性强的无量纲特征,降低分类模型复杂度并提高算法速度;最后通过选取合适的SVM分类模型进行分类诊断。结合具有无量纲特征的故障敏感性与SVM的非线性分类性进行诊断分类,并通过石化机组故障诊断实验平台进行验证,表明该方法相比于其他经典分类方法分类效果更好,分类正确率为99.1%,证明了方法的有效性。

固有成分滤波器的旋转机械故障诊断方法

针对噪声环境下旋转机械微弱复合故障诊断问题,提出了一种强噪声干扰下基于固有成分滤波器(intrinsic component filtering,简称ICF)的旋转机械故障检测和分离方法。ICF通过最小化样本间特征的L1/2范数和样本内特征的L3/2范数来实现样本之间特征的一致性和样本内部特征的稀疏性,并训练出最优滤波器组,是一种无监督多维盲解卷积算法。首先,构建输入信号的Hankel训练矩阵,通过权值矩阵与Hankel矩阵的乘积模拟卷积过程,再利用固有属性滤波器实现特征学习;其次,通过峭度信息选择最优滤波器;最后,根据滤波后的时域波形和包络谱实现故障诊断。仿真和试验信号验证了提出方法的故障诊断性能,研究结果表明,提出的方法无需任何先验经验,可以实现强噪声环境下的微弱故障的分离,同时具备很好的鲁棒性。

基于多源域数据与机器学习算法的转子不平衡故障诊断

国内能源生产装置规模大型化发展趋势明显,与其配套的旋转机械设备发生故障导致的非计划停机将会造成严重的经济损失与重大安全问题。转子不平衡贯穿了旋转机械设备的整个生命周期,服役转子的状态诊断格外重要。针对大型旋转机械振动测点较多,振动信号具有非平稳特征等问题,提出基于多源域数据提取与机器学习算法的转子不平衡故障诊断模型。首先以多源振动监测数据为驱动,根据互相关系数提取故障信息丰富的振动信号,融合时域、频域、时频域等多域特征构建高维混合特征空间;其次利用基于t分布的随机邻域嵌入方法揭示高维空间的特征信息,反映为可视化的三维空间;最终通过最邻近节点算法进行故障分类,判断转子的不平衡质量与相位。本文提出利用互相关系数表征多源数据的故障信息丰富程度,并结合机器学习手段判断转子不平衡类型。通过设计不同附加质量的转子在多转速下不平衡状态实验,验证了所提模型的有效性,解决了转子在线诊断和现场动平衡问题。