基于孤立森林算法的风电齿轮箱轴承故障检测方法

为精准检测齿轮箱轴承故障,实时进行有效监测以保证风电齿轮箱健康运行,提出一种基于孤立森林算法的风电齿轮箱轴承故障检测方法.首先,以齿轮箱轴承温度为故障检测模型的输出变量,采用多尺度图相关算法选择输入变量;然后,提取输入变量的均方根和包络线进行自组织映射神经网络特征融合;最后,以融合值为模型输入量,使用孤立森林算法进行异...     

Isolability of faults in sensor fault diagnosis

A major concern with fault detection and isolation (FDI) methods is their robustness with respect to noise and modeling uncertainties. With this in mind, several approaches have been proposed to minimize the vulnerability of FDI methods to these uncertainties. But, apart from the algorithm used, there is a theoretical limit on the minimum effect of noise on detectability and isolability. This limit has been quantified in this paper for the problem of sensor fault diagnosis based on direct redundancies. In this study, first a geometric approach to sensor fault detection is proposed. The sensor fault is isolated based on the direction of residuals found from a residual generator. This residual generator can be constructed from an input-output or a Principal Component Analysis (PCA) based model. The simplicity of this technique, compared to the existing methods of sensor fault diagnosis, allows for more rational formulation of the isolability concepts in linear systems. Using this residual generator and the assumption of Gaussian noise, the effect of noise on isolability is studied, and the minimum magnitude of isolable fault in each sensor is found based on the distribution of noise in the measurement system. Finally, some numerical examples are presented to clarify this approach.     

Intelligent measurement methods in technical diagnosis and quality assurance - a comparison

Faults in technical systems cannot be measured directly. Thus, a variety of intelligent measurement methods has been developed to detect faults. Two widely used procedures-the vibration analysis and the parameter estimation method, and their applications to technical diagnosis - will be described and compared.The paper deals with the application of the vibration analysis to the fault detection in a gear and the application of the parameter estimation to the fault detection in an electric drive system. The paper shows that faults in auxiliary systems canbe detected better by the vibration analysis, whereas faults in the drive system can be detected by the parameter estimation method. A combination of both methods provides better test results than conventional procedures.     

Weak feature enhancement in machinery fault diagnosis using empirical wavelet transform and an improved adaptive bistable stochastic resonance

Machinery vibration signal is a typical multi-component signal and fault features are often submerged by some interference components. To accurately extract fault features, a weak feature enhancement method based on empirical wavelet transform (EWT) and an improved adaptive bistable stochastic resonance (IABSR) is proposed. This method makes full use of the signal decomposition performance of EWT and the signal enhancement of the IABSR to achieve the purpose of fault feature enhancement in low frequency band of FFT spectrum. Firstly, EWT is used as the preprocessing program of bistable stochastic resonance (BSR) to decompose the machinery vibration signal into a set of sub-components. Then, the sensitive component that contains main fault information is further input into BSR system to enhance fault features with the assistance of residual noises. Finally, the fault features are identified from fast Fourier transform (FFT) spectrum of the BSR output. To achieve the optimal BSR output, the IABSR method based on salp swarm algorithm (SSA) is presented. Compared with the tradition adaptive BSR (ABSR), the IABSR optimizes not only the BSR system parameters but also the calculation step size. Two case studies on machinery fault diagnosis demonstrate the effectiveness and superiority of the proposed method. In addition, the proposed method is easy to implement and is robust to noise to some extent.     

Robust fault detection for the dynamics of high-speed train with multi-source finite frequency interference

This paper proposes a composite fault detection scheme for the dynamics of high-speed train (HST), using an unknown input observer-like (UIO-like) fault detection filter, in the presence of wind gust and operating noises which are modeled as disturbance generated by exogenous system and unknown multi-source disturbance within finite frequency domain. Using system input and system output measurements, the fault detection filter is designed to generate the needed residual signals. In order to decouple disturbance from residual signals without truncating the influence of faults, this paper proposes a method to partition the disturbance into two parts. One subset of the disturbance does not appear in residual dynamics, and the influence of the other subset is constrained by $H_{\infty }$ performance index in a finite frequency domain. A set of detection subspaces are defined, and every different fault is assigned to its own detection subspace to guarantee the residual signals are diagonally affected promptly by the faults. Simulations are conducted to demonstrate the effectiveness and merits of the proposed method.     

带窗口平移鲁棒故障及其在水轮发电机监控中的应用

针对一水轮发电机单机模型进水闸门位置故障,用对等空间法检测了未知输入干扰条件下的系统故障,为了减少计算量,提高计算速度,采用窗口平移的残差计算方法。结果表明这种计算方法十分有效,残差对故障非常敏感,而对噪声具有很强的鲁棒性。与文献［４］辨识法给出的结果比较,对等空间法的残差对故障的响应具有持续性,而不是瞬态性,有利于准确检测故障,降低故障误报率。     

基于WA-ESN的建筑起重机械故障检测

为了提高建筑起重机械故障检测精度,提出了小波消噪和回声状态网络(ESN)的建筑起重机械故障检测方法(WA-ESN)。首先,采用小波分析(WA)去除建筑起重机械故障振动信号的噪声,并提取故障部件振动号故障特征,然后,将特征向量作为回声状态网络的输入向量,故障类型作为输出,进行训练,建立建筑起重机械故障智能检测模型。测试结果表明,该方法提高了建筑起重机械故障检测精度,减少了建筑起重机械故障检测检测误差,具有较高的实际应用价值。     

Design of H(infinity) robust fault detection filter for linear uncertain time-delay systems

In this paper, the robust fault detection filter design problem for linear time-delay systems with both unknown inputs and parameter uncertainties is studied. Using a multiobjective optimization technique, a new performance index is introduced, which takes into account the robustness of the fault detection filter against disturbances and sensitivity to faults simultaneously. The reference residual model is then designed based on this performance index to formulate the robust fault detection filter design problem as an H(infinity) model-matching problem. By applying robust H(infinity) optimization control technique, the existence condition of the robust fault detection filter for linear time-delay systems with both unknown inputs and parameter uncertainties is presented in terms of linear matrix inequality formulation, independently of time delay. In order to detect the fault, an adaptive threshold which depends on the inputs is finally determined. An illustrative design example is used to demonstrate the validity of the proposed approach.     

A modified NARMAX model-based self-tuner with fault tolerance for unknown nonlinear stochastic hybrid systems with an input–output direct feed-through term

A modified nonlinear autoregressive moving average with exogenous inputs (NARMAX) model-based state-space self-tuner with fault tolerance is proposed in this paper for the unknown nonlinear stochastic hybrid system with a direct transmission matrix from input to output. Through the off-line observer/Kalman filter identification method, one has a good initial guess of modified NARMAX model to reduce the on-line system identification process time. Then, based on the modified NARMAX-based system identification, a corresponding adaptive digital control scheme is presented for the unknown continuous-time nonlinear system, with an input–output direct transmission term, which also has measurement and system noises and inaccessible system states. Besides, an effective state space self-turner with fault tolerance scheme is presented for the unknown multivariable stochastic system. A quantitative criterion is suggested by comparing the innovation process error estimated by the Kalman filter estimation algorithm, so that a weighting matrix resetting technique by adjusting and resetting the covariance matrices of parameter estimate obtained by the Kalman filter estimation algorithm is utilized to achieve the parameter estimation for faulty system recovery. Consequently, the proposed method can effectively cope with partially abrupt and/or gradual system faults and input failures by the fault detection.     

Robust unknown input observer based fault detection for high-order multi-agent systems with disturbances

This paper is devoted to fault detection (FD) for high-order multi-agent systems with disturbances. In order to detect the fault occurred in one agent, the unknown input observer (UIO) is constructed in its neighbor. Two cases are considered, if the perfect UI decoupling condition is satisfied, the UI does not affect the residual; if the condition is not satisfied, this paper proposes a method of partitioning the UI into two parts, such that a subset of the UI does not appear in residual dynamics, and the influence of the other UI is constrained. Simulations are given to demonstrate the effectiveness of the proposed method.     

A study on fault diagnosis and maintenance of CNC-WEDM based on binary relational analysis and expert system

The paper presents a binary relational analysis and expert system base module for maintenance and fault diagnosis of CNC wire EDM. The module proposes a framework of integrated maintenance and fault diagnosis system. The study explores the binary coded matrix system, which plays an important role in prediction and diagnosis of wire electrical discharge machining (WEDM) faults on the spot by expert guidance. In this study, 15 inputs were considered to observe eight probable causes with the help of the forward and backward propagation algorithms. Inputs and output matrices were considered in the form of a square matrix. To explain the fault diagnosis and to realize the importance of maintenance through advice, the detection of faults is investigated through forward and back propagation of matrix transformation on the spot. It is an integrated backup that can be individually focused when input and output parameter do not match. It is a time saving, knowledge acquisition, easy to maintain, and capable of self-learning system. To verify the developed framework, 120 data sets were generated for proper analyzing of acquired output through graphical representation. The paper also presents some of the important features of maintenance schedule and probable causes of wire breakage with remedial actions in tabular form. The developed system can help the operators, trainees, and manufacturing engineers in achieving trouble free machining through quick detection of faults and proper maintenance of machines in actual practice.     

A study on fault diagnosis and maintenance of CNC-WEDM based on binary relational analysis and expert system

The paper presents a binary relational analysis and expert system base module for maintenance and fault diagnosis of CNC wire EDM. The module proposes a framework of integrated maintenance and fault diagnosis system. The study explores the binary coded matrix system, which plays an important role in prediction and diagnosis of wire electrical discharge machining (WEDM) faults on the spot by expert guidance. In this study, 15 inputs were considered to observe eight probable causes with the help of the forward and backward propagation algorithms. Inputs and output matrices were considered in the form of a square matrix. To explain the fault diagnosis and to realize the importance of maintenance through advice, the detection of faults is investigated through forward and back propagation of matrix transformation on the spot. It is an integrated backup that can be individually focused when input and output parameter do not match. It is a time saving, knowledge acquisition, easy to maintain, and capable of self-learning system. To verify the developed framework, 120 data sets were generated for proper analyzing of acquired output through graphical representation. The paper also presents some of the important features of maintenance schedule and probable causes of wire breakage with remedial actions in tabular form. The developed system can help the operators, trainees, and manufacturing engineers in achieving trouble free machining through quick detection of faults and proper maintenance of machines in actual practice.     

Classifying flexible measures in data envelopment analysis: A slack-based measure

Data envelopment analysis (DEA) has gained a wide range of applications in measuring comparative efficiency of decision making units (DMUs) with multiple incommensurate inputs and outputs. The standard DEA method requires that the status of all input and output variables be known exactly. However, in many real applications, the status of some measures is not clearly known as inputs or outputs. These measures are referred to as flexible measures. This paper proposes a flexible slacks-based measure (FSBM) of efficiency in which each flexible measure can play input role for some DMUs and output role for others to maximize the relative efficiency of the DMU under evaluation. Further, we will show that when an operational unit is efficient in a specific flexible measure, this measure can play both input and output roles for this unit. In this case, the optimal input/output designation for flexible measure is one that optimizes the efficiency of the artificial average unit. An application in assessing UK higher education institutions used to show the applicability of the proposed approach.     

基于幂函数型双稳随机共振的故障信号检测方法

在实际的故障诊断中,有用信号经常淹没在噪声中,特征信息提取非常困难。为了提取强噪声背景中的微弱信号,将幂函数型单势阱模型与Gaussian Potential模型相结合提出一种新型的双稳随机共振系统,称为幂函数型双稳随机共振系统。首先,以平均信噪比增益为衡量指标,提出一种寻找最优系统参数组合的算法,使微弱信号、噪声及系统产生最佳的共振效果;然后,基于幂函数型双稳随机共振系统对Levy噪声背景下的仿真信号进行检测;最后提出一种基于小波变换和幂函数型双稳随机共振的微弱信号检测方法并应用于轴承故障信号检测中。仿真实验表明,幂函数型双稳随机共振模型在故障信号检测中是有效和可靠的。     

Detection of minute signs of a small fault in a periodic or a quasi-periodic signal by the harmonic wavelet transform

If a machine in operation has a fault, signs of the fault appear in the monitored signal and are usually synchronised with the operating speed. The signs are very small when the fault is at an early stage. The fast Fourier transform (FFT) is often utilised to detect these signs, but it is very difficult to detect minute signs. In this paper, harmonic wavelet transform is utilised to detect the minute signs of small faults in a monitored signal. The monitored signal of a machine element, in ordinary operation, is regarded as periodic or quasi-periodic. It is important for the effectual detection of the minute signs to reduce the obstructive noise and the end effects in the signal. The end effect is a peculiar phenomenon to wavelet transform and hampers effective detection. Useful methods to reduce the obstructive noise and the end effects are devised herein by the authors. Even if no visible information pertaining to a fault appears in the monitored waveform, the present method is able to detect a minute sign of a small fault. It is demonstrated that the present method is capable of detecting minute signs arising from slight collisions caused by a loose coupling and a fatigue crack.     

基于粒子滤波与负向选择算法的GIS设备故障检测

为了及早发现故障合理安排设备检修计划,提出一种基于粒子滤波与负向选择算法的GIS故障检测方法。首先,选取GIS设备金属外壳振动信号分形维数作为特征变量,有效削弱了设备负载变化对外壳振动的影响。同时,基于粒子滤波及支持向量回归算法处理设备正常状态下的振动信号分形维数特征样本,建立GIS设备振动特征估计器。将实时测量的振动特征输入特征估计器,计算估计器输入值与输出值之间的残差并作为检测指标。最后,利用负向选择算法处理正常状态下检测指标数据,间接获取GIS故障状态下检测指标区间,进而实现设备故障的检测。通过对现场实际测量数据的仿真分析,验证了所提方法的可行性。     

基于互相关检测的滚动轴承实时故障诊断方法

提出了一种易于用模拟电路实现的基于互相关检测的滚动轴承实时故障诊断方法,首先,用两个加速度传感器在不同测点采集轴承振动信号,将其分别送入相应通道的高Q带通滤波器来选择最优共振带;然后,将两路带通滤波器的输出信号进行互相关检测,将互相关检测得到的信号经低通滤波器,保留低频故障信号;最后,将低通滤波器输出的时域信号通过频谱分析仪显示滚动轴承故障特征频率的谱线以实现滚动轴承的实时故障诊断。用模拟电路的形式将该方法进行搭建,并在QPZZ-II实测平台完成滚动轴承的实时故障诊断。结果表明:该方法克服了单一信号源的局限性,能利用互相关函数削弱共振带内部噪声,使诊断结果具有更高的频谱辨识率,而且能够用结构简单、易于维护的模拟电路实现,对轴承实时故障诊断方法的应用与普及具有一定的参考价值。     

Lyapunov指数算法在应答器信号混沌振子检测中的应用

针对为提高在强噪声环境下应答器上行链路传输信号的检测精度,利用混沌系统对初始条件敏感以及对噪声免疫的特性,将混沌振子应用到应答器上行链路信号检测解调中.结合微弱信号Duffing振子检测原理和应答器上行链路信号特征,给出了使用Duffing振子检测应答器信号的方法和步骤,并使用Lyapunov指数算法计算Duffing振子检测系统的临界阈值,定量判断系统的输出状态,实现应答器信号的解调.在理论分析的基础上,进行了实验仿真验证.仿真结果表明,基于Lyapunov指数算法的应答器信号混沌振子检测方法提高了阈值设置的准确性和效率,并确保了应答器信号检测的可靠性.     

在轨飞轮故障诊断混合框架设计

为提升飞轮的可靠性,本文对飞轮故障诊断技术进行了研究。通过对基于数学解析模型与基于智能计算的故障诊断方法的对比研究,提出了一种基于神经网络的混合故障诊断方法。该方法首先使用数学解析模型与原系统输出的差值作为一级残差;而后利用该一级残差以及系统可测状态对神经网络进行训练;然后使用混合模型输出的二级残差对系统故障进行检测;最后以飞轮注入母线电压以及电枢电流故障对该方法进行验证:在存在母线电压故障工况下混合模型避免了解析模型电流估计的发散问题,与单神经网络模型相比最大跟踪误差降低了44%。在存在电流故障时,不同的转速工况下与两种单模型相比混合模型的最大跟踪误差降低了90%,跟踪方差减小了10倍以上。混合方法可以有效解决由于解析模型存在建模误差引起的故障诊断不够准确的问题以及由于缺乏训练数据所引起的单神经网络模型不能适应新工况的故障诊断问题。