基于阶比分析的转子系统油膜涡动故障诊断研究

针对转子升速过程中油膜涡动信号的非平稳特点,将阶比分析应用于油膜涡动故障诊断中。首先对转子升速时的振动信号进行时域采样,再对时域信号进行等角度采样,将其转化为角域平稳信号,然后对角域信号进行频谱分析识别油膜涡动故障频率。对实测油膜涡动信号分析表明,阶比分析能有效地避免传统频谱分析方法中的"频率模糊"现象,能够准确识别转子升速过程中油膜涡动故障频率。     

基于EEMD-LSTM的汽轮机转子碰磨故障诊断模型及其工程应用

汽轮机转子与静子间的碰磨严重影响着机组的安全运行。为了解决汽轮机转子发生在早期和中期的碰磨故障难以通过基于振动信号检测诊断方法进行有效识别的问题,本文提出一种基于EEMD-LSTM的汽轮机转子碰磨故障诊断方法。首先,该方法通过声发射技术监测汽轮机转子的碰磨故障信号;然后,利用EEMD信号分解方法处理获取的声发射信号,并提取能量特征参数和相关的时域特征参数,从而获得碰磨故障特征数据集;最后,利用划分的数据集对LSTM神经网络进行训练与测试,从而获得碰磨故障诊断模型。工程应用结果表明,本文提出的方法能够有效识别机组在不同转速时期的早期碰磨故障,且故障诊断的准确率较高。     

A novel wind turbine health condition monitoring method based on common features distribution adaptation

Aimed at the difficulty of diagnosing the transmission system of wind turbine under variable working conditions, a novel health condition monitoring method based on common features distribution adaptation is proposed in this article. In the method, envelope analysis is first performed on the collected signals, and then the time-frequency features are extracted to be combined as new input samples. The feature set under the working condition similar to target working condition is selected as the auxiliary sample set in source domain through the evaluation of the transferability. The kernel function is used to map the labeled auxiliary samples and unlabeled target samples to a reproduced kernel Hilbert space, which effectively reduces the data distribution discrepancy between source and target domains. The problem of class imbalance in each domain is taken into account when performing fault recognition, which improves the effect of transfer learning. Finally, the adjusted source domain is used to train the classifier, which is applied to the target domain to get the predicted labels of the test data. Experiment shows that the proposed method has better working performance than traditional fault diagnosis methods.     

Fault detection and diagnosis methods for green hydrogen production: A review

One of the green hydrogen projects is Zero Emission Hydrogen Turbine Center (ZEHTC), in which solar panels, PEM electrolyzer, and diaphragm compressor are used to generate power, produce hydrogen and store hydrogen at high pressure, respectively. Faults in any components of photovoltaic (PV) systems, PEM electrolyzers, and diaphragm compressors can seriously affect the efficiency, energy yield as well as security, and reliability of the entire system, if not detected and corrected quickly. In this paper, the types and causes of PV systems, PEM electrolyzer, and diaphragm compressors failures are presented, then different methods proposed in the literature for fault detection and diagnosis (FDD) of systems are reviewed and discussed. Special attention is paid to methods that can accurately detect, localize and classify possible faults occurring in a PV arrays. The advantages and limits of FDD methods in terms of feasibility, complexity, cost-effectiveness and generalization capability for large-scale integration are highlighted. Based on the reviewed papers, challenges and recommendations for future research direction are also provided. In this work different model-based approaches are investigated as well as their validation and applications. An overview of different methodologies available in the literature is proposed, which is oriented to help in developing suitable diagnostic tool for PEM electrolyzer monitoring and fault detection and isolation (FDI). Model-based methods provide fault detection and identification, are easy to implement, and could be conducted during system operation.     

Acoustic emission-based damage characterization of 70 MPa type IV hydrogen composite pressure vessels during hydraulic tests

This paper aims to characterize the damage mechanisms of 70 MPa Type IV hydrogen composite pressure vessels using the acoustic emission (AE) method. First, AE signals were captured during the 0–105 MPa and 0–158 MPa hydraulic tests of two vessels using multi-step loading method. Second, the AE feature parameters in time-domain and frequency-domain such as amplitude, frequency, and energy are studied. A multi-parameter statistical analysis (MPSA) method based on empirical mode decomposition (EMD) and K-means algorithm is performed to cluster AE events for the vessels. Intrinsic mode functions (IMFs) are decomposed by EMD and three IMFs with high frequency are chosen to reconstruct the feature parameters and provide signal pre-processing for K-means clustering analysis. Based on the relationship between AE features and damage modes, three main clusters with separate amplitude, absolute energy, and energy are correlated to matrix cracking, fiber/matrix debonding, and fiber breakage damage mechanisms. Besides, the effectiveness of MPSA method for signal classification is validated by principal component analysis (PCA) and fast Fourier transformation (FFT) method. Finally, the AE feature parameters such as amplitude and counts to peak for the three main damage modes are studied for the hydraulic proof tests and the burst tests to explore the damage evolution behaviors of the vessels with pressure increasing. Results show that AE method can be reliably used to characterize damage evolution mechanisms in composite pressure vessels.     

基于优化VMD复合多尺度散布熵及LSTM的风力发电机齿轮箱故障诊断方法研究

以风力发电机齿轮箱加速度信号为研究对象,提出一种数据驱动的风力发电机齿轮箱故障诊断方法,该方法以灰狼优化的变分模态分解方法(AGWO-VMD)、复合多尺度规范化散布熵(NCMDE)及长短期记忆网络(LSTM)为基础,实现齿轮箱故障的快速诊断.首先将时域信号转换至角域;然后通过AGWO-VMD方法对角域信号进行自适应分解...     

基于EEMD及分形方法的轴承振动信号分析

为对具有强烈非线性特征的轴承振动信号做出准确的故障识别,基于分形理论,采用辅助经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)对信号进行降噪预处理,采用G-P算法分析轴承不同状态下振动信号关联维数。研究表明:基于EEMD的降噪方法可有效对振动信号进行降噪;轴承工作状态不同,其振动信号关联维数具有明显的可区分性,当轴承处于外环故障时,其关联维数最大为4.7,当轴承处于滚珠故障时,其关联维数最小仅为3.0,当轴承处于正常/内环故障时,其关联维数分别为4.0/3.2。因此,利用关联维数能定量识别轴承的不同故障状态及位置。     

基于S-SLLE的风电机组齿轮箱故障诊断方法研究

针对风电机组齿轮箱结构复杂、受交变载荷和恶劣工作环境影响容易出现故障导致停机的问题,提出基于统计学K-均值聚类理论的统计型监督式局部线性嵌入流形学习(S-SLLE)特征维数约简方法,首先通过对齿轮箱振动信号时频域故障特征提取,剔除冗余特征向量,减少诊断模型的复杂度和计算量,再利用RBF核支持向量机分类器建立诊断模型,对...     

基于EEMD的变压器振动与声音信号IMF峭度特征量提取方法

针对变压器振动问题,提出了一种基于集合经验模态分解(EEMD)的本征模函数(IMF)峭度特征量提取方法,并运用相关系数法、快速谱峭度图法提取敏感IMF分量。提取试验变压器正常、铁心松动故障状态下的振动、声音信号的特征量,研究变压器在正常、故障状态下这两种信号特征量分布情况;分析实际运行中出现铁心磁路故障、铁心多点接地故障状态的变压器的IMF峭度特征。结果表明,提出的特征量提取方法可同时反映频域、时域特性;在不同故障条件下,振动与声音信号的特征量变化不同,二者可相互补充,研究两种信号更有利于变压器状态的判定。     

A new cavity profile for a diaphragm compressor used in hydrogen fueling stations

The short operating life of metallic diaphragm caused by fracture is one of the main disadvantages for diaphragm compressors used in hydrogen fueling stations. A new generatrix for cavity profile is proposed through optimization using the complex method to decrease the maximal radial stresses on both oil and gas sides of the diaphragm clinging to the cavity surface. In the optimization, the convex part of the cavity generatrix is subjected to a constraint that the generatrix has a lower slope than the deformed diaphragm under a uniform pressure load. This constraint aims to avoid cavity dead volume at the end of the gas discharge process. Thus, an analytical solution for the deflection of an edge-clamped metallic diaphragm under a uniform pressure load is firstly developed. The solution employs the principle of minimum energy and the Rayleigh-Ritz method, which based on the theory of thin plates with large deflections. Experimental measurements, as well as the finite elements method (FEM), are employed to validate the solution. The analytical results are found to be in good agreement with the results of measurements and FEM simulations. Secondly, the stress of the diaphragm with a specific deflection is calculated, and the radial stress concerning both gas side and oil side of the diaphragm is taken as the objective function. Finally, a new generatrix is obtained through the optimization. The radial stress of the diaphragm clinging to the new cavity profile is validated via the FEM simulation, and results match well with each other. It also approves that the cavity dead volume is eliminated by the new generatrix at the end of the gas discharge process. Moreover, the maximal and the centric radial stress of the working diaphragm were compared between the new generatrix and the traditional generatrix under the same design parameters, the maximal and the centric radial stress of the diaphragm decreased by 8.2% and 13.9%, respectively. Based on the proposed design method, effects of the cavity volume, cavity radius, diaphragm thickness and diaphragm material properties on the maximal radial stress of the working diaphragm are further discussed.     

Diagnostic of fuel cell air supply subsystems based on pressure signal records and statistical pattern recognition approach

A data-driven and application-oriented diagnosis tool is developed for Fuel Cell (FC) air supply subsystems. A bench emulating a FC air line is built to study normal and abnormal operations (clogged inlet, air leakage, error in compressor speed control) and data are collected using the air pressure transducer, which is usually implemented in FC generators. A pattern recognition approach is then applied to statistical features extracted from the pressure signal. The performance of the diagnosis strategy is evaluated from confusion matrices, associated to graphs and performance indicators. Two examples of compressors, air subsystem managements, and data records are considered to examine the method portability. Best classification rates (>95%) are obtained on test profiles, when the pressure regulation is disabled; fault stamps can thus be found in the pressure signal morphology. Regarding the frequency of data logging, both 1 kHz and 100 Hz values are found effective for fault isolations.     

基于变分模态分解和去趋势波动分析的柴油机振动信号去噪方法

为了解决柴油机工作时其振动信号的背景噪声对状态监测及故障诊断造成干扰这一问题,提出一种基于变分模态分解(VMD)和去趋势波动分析(DFA)的柴油机振动信号去噪方法。该方法首先利用变分模态分解将振动信号分解为若干分量,再利用去趋势波动分析分别计算各个分量的尺度指数,根据尺度指数的值选取具有长程相关性的分量进行信号的重构,以消除振动信号中噪声。将该方法应用于仿真信号和柴油机故障振动信号中,取得了良好的消噪效果。     

Diagnosis method based on wavelet coefficient scale relativity correlation dimension for fault

Correlation dimension as a tool to describe machinery condition is introduced. Vibration signals of the fan under different working conditions are analyzed using a threshold filtering algorithm based on the region relativity of the wavelet coefficients for reducing noise. The result shows that the characteristics of the signal could be preserved completely. The correlation dimension is able to identify conditions of the fan with faults compared with the normal condition, thereby providing an effective technology for condition monitoring and fault diagnosis of mechanical equipment.     

一种W型结构元素的自适应数学形态学及其在风电机组轴承故障诊断中应用

为了提高风电机组滚动轴承故障诊断的有效性和可靠性,提出一种W型自适应数学形态学特征提取方法,并与谱相关分析相结合形成风电机组滚动轴承故障诊断策略。该方法首先针对传统三角型结构元素在故障特征提取中易出现对脉冲信号的漏查,提出一种W型结构元素,旨在捕捉更多特征信息;之后依据各故障信号的实际波形得到结构元素的高和最优开闭运算加权因子,构建自适应形态学模型;最后对测试信号与训练信号进行频域内谱相关性分析,依据相关系数识别故障。将该方法通过数值例、西储大学实验台轴承数据和实际风场采集数据进行算法验证,并与传统的三角型结构元素进行比较,实验结果表明W型结构元素能更有效地提取信号中的脉冲成分、降低噪声干扰,故障诊断算法可准确识别出故障类别,提高结果的可靠性。     

机组状态监测与故障诊断系统在紧水滩水电厂的应用

介绍紧水滩水电厂机组状态监测系统的构成，分析振动摆度监测、水力压力脉动监测、水轮机效率监测、水轮机汽蚀监测的技术特点，以及智能故障诊断系统的基本结构。     

互补集合经验模式分解与奇异值能量谱在风电齿轮故障识别中的应用

针对风电机组齿轮系统故障模式的有效识别问题,提出一种互补集合经验模式分解(CEEMD)与奇异值能量谱相结合的故障识别方法。利用CEEMD将齿轮非平稳信号分解为有限个平稳的本征模态函数,并将其组成初始特征向量矩阵,对矩阵进行奇异值分解并求出风电齿轮不同工况下的奇异值能量谱分布,以奇异值能量谱为元素构造特征向量,通过计算不同工况振动信号的灰色关联度来判断齿轮的故障类型。实例表明,该方法能有效应用于风电机组齿轮系统的故障诊断。     

基于AR模型和K-L信息量的柴油机气阀机构故障诊断

通过模拟气阀机构的两种常见故障:气阀漏气和气阀间隙异常,采集柴油机缸盖表面的振动信号.提出了柴油机气阀机构的状态监测及故障诊断策略,采用FPE准则和Burg算法建立不同状态时振动信号的AR模型,利用K-L信息量对不同工作状态进行了有效识别.诊断结果表明该方法是可行的,便于实现柴油机气阀机构故障的在线实时监测与诊断.     

Effect of hydraulic oil compressibility on the volumetric efficiency of a diaphragm compressor for hydrogen refueling stations

The low volumetric efficiency of the diaphragm compressor under hydrogen refueling process, which hereby results in poor energy efficiency and high cost of hydrogen applications, should be paid attention to. This paper presents theoretical analysis and experimental investigation of the factors affecting the volumetric efficiency of the diaphragm compressor for hydrogen refueling process, focusing on the influence of hydraulic oil compressibility. A mathematical model was established to estimate the volumetric efficiency of diaphragm compressors, in which the effects of clearance volume, superheating of suction gas and pressure loss were taken into account and the emphasis was focused on the compressibility of hydraulic oil. A test rig was built to validate the theoretical model and further experimental investigations were carried out to identify the factors influencing the oil compressibility and hereby the volumetric efficiency. The volumetric efficiency was measured and compared under varied oil compressibility conditions by varying elastic modulus, oil overflow pressure and oil volume. The results indicated that the measured volumetric efficiency agrees well with the calculated value. The compression and expansion of hydraulic oil have a dominant influence on the volumetric efficiency, resulting in a loss of 37% of volumetric efficiency as compared to 2.4%, 18% and 1%, respectively for losses associated with clearance volume, superheating of suction gas and pressure loss, for a diagram compressor under refueling conditions with suction pressure of 30 MPa and discharge pressure of 90 MPa. The volumetric efficiency reduced rapidly with the increased oil overflow pressure, at a rate of 5% decrease with every 10 MPa rise in oil overflow pressure. As the oil volume increased by 100% of the stroke volume, the volumetric efficiency droped by 5.5%.     

基于EMD与SVM的柴油机故障诊断

对柴油机气阀间隙变化、断油等故障情况下的缸盖振动信号进行了测试分析.采用经验模式分解EMD方法对振动信号进行分解,得到固有模态函数IMF,对每一个IMF分量分别建立AR模型,以模型的自回归参数和残差的方差作为特征向量,用支持向量机SVM进行分类,判断柴油机的工作状态和故障类型.实验结果分析表明,该方法即使在小样本情况下也能准确有效地诊断柴油机故障,能实现故障的实时自动化诊断.     

Generator bearing fault diagnosis for wind turbine via empirical wavelet transform using measured vibration signals

The implementation of condition monitoring and fault diagnosis system (CMFDS) on wind turbine is significant to lower the unscheduled breakdown. Generator is one of the most important components in wind turbine, and generator bearing fault identification always draws lots of attention. However, non-stationary vibration signal of weak fault and compound fault with a large amount of background noise makes this task challenging in many cases. So, effective signal processing method is essential in the accurate diagnosis step of CMFDS. As a novel signal processing method, empirical Wavelet Transform (EWT) is used to extract inherent modulation information by decomposing signal into mono-components under an orthogonal basis, which is seen as a powerful tool for mechanical fault diagnosis. Moreover, in order to avoid the inaccurate identification the internal modes caused by the heavy noise, wavelet spatial neighboring coefficient denoising with data-driven threshold is applied to increase Signal to Noise Ratio (SNR) before EWT. The effectiveness of the proposed technique on weak fault and compound fault diagnosis is first validated by two experimental cases. Finally, the proposed method has been applied to identify fault feature of generator bearing on wind turbine in wind farm successfully.