Double-layer ensemble monitoring of non-gaussian processes using modified independent component analysis

The modified independent component analysis (MICA) was proposed mainly to obtain a consistent solution that cannot be ensured in the original ICA algorithm and has been widely investigated in multivariate statistical process monitoring (MSPM). Within the MICA-based non-Gaussian process monitoring circle, there are two main problems, i.e., the selection of a proper non-quadratic function for measuring non-Gaussianity and the determination of dominant ICs for constructing latent subspace, have not been well attempted so far. Given that the MICA method as well as other MSPM approaches are usually implemented in an unsupervised manner, the two problems are always solved by some empirical criteria without respect to enhancing fault detectability. The current work aims to address the challenging issues involved in the MICA-based approach and propose a double-layer ensemble monitoring method based on MICA (abbreviated as DEMICA) for non-Gaussian processes. Instead of proposing an approach for selecting a proper non-quadratic function and determining the dominant ICs, the DEMICA method combines all possible base MICA models developed with different non-quadratic functions and different sets of dominant ICs into an ensemble, and a double-layer Bayesian inference is formulated as a decision fusion method to form a unique monitoring index for online fault detection. The effectiveness of the proposed approach is then validated on two systems, and the achieved results clearly demonstrate its superior proficiency.     

A new thermographic NDT for condition monitoring of electrical components using ANN with confidence level analysis

Infrared thermography technology is one of the most effective non-destructive testing techniques for predictive faults diagnosis of electrical components. Faults in electrical system show overheating of components which is a common indicator of poor connection, overloading, load imbalance or any defect. Thermographic inspection is employed for finding such heat related problems before eventual failure of the system. However, an automatic diagnostic system based on artificial neural network reduces operating time, human efforts and also increases the reliability of system. In the present study, statistical features and artificial neural network (ANN) with confidence level analysis are utilized for inspection of electrical components and their thermal conditions are classified into two classes namely normal and overheated. All the features extracted from images do not produce good performance. Features having low performance reduce the diagnostic performance. The study reveals the performance of each feature individually for selecting the suitable feature set. In order to find the individual feature performance, each feature of thermal image was used as input for neural network and the classification of condition types were used as output target. The multilayered perceptron network using Levenberg–Marquardt training algorithm was used as classifier. The performances were determined in terms of percentage of accuracy, specificity, sensitivity, false positive and false negative. After selecting the suitable features, the study introduces the intelligent diagnosis system using suitable features as inputs of neural network. Finally, confidence percentage and confidence level were used to find out the strength of the network outputs for condition monitoring. The experimental result shows that multilayered perceptron network produced 79.4% of testing accuracy with 43.60%, 12.60%, 21.40, 9.20% and 13.40% highest, high, moderate, low and lowest confidence level respectively.     

Condition monitoring of steam turbine-generator through contamination analysis of used lubricating oil

The objective of this paper is to present a systematic analysis of wear particles contained in used lubricant of steam turbine-generator of a thermal power station. The turbogenerator was condition-monitored over a period of two years through wear debris and particulate contamination analysis of the oil. Various sophisticated techniques such as automatic particle counter, ferrography, inductively coupled plasma atomic emission spectrometry (ICPAES), scanning electron microscopy and energy dispersive X-ray analysis (SEM/EDAX) have been employed to extract the relevant information about the health of the machine. Eventually, a correlation of different techniques of wear debris monitoring on the basis of current investigation ascertains the significance of the collective approach of various techniques to avoid catastrophic breakdowns and expensive component replacements.     

Analytical approach to wear rate determination for internal combustion engine condition monitoring based on oil analysis

Wear has important, negative effects on the functioning of engine parts. Additionally, this situation is very difficult to evaluate accurately in oil analysis for engine condition monitoring. Original Equipment Manufacturers (OEM), lubricant suppliers and oil analysis laboratories provide specific guidelines for wear metal concentrations. These limits provide good general guidelines for interpreting oil analysis data, but do not take into account common factors that influence the concentration of wear debris and contaminants in an oil sample. These factors involve oil consumption, fresh oil additions, etc., and particular features such as engine age, type of service, environmental conditions, etc.In this paper, an analytical approach to enable a more accurate wear determination from engine oil samples is developed. The above factors are taken into account and an improved maintenance program for internal combustion engines based on oil analysis is developed.     

Condition monitoring of aircraft by quantitative filter debris analysis

Various condition monitoring techniques are used collectively to monitor the health of aircraft engines and transmission, a concept known as Integrated Health Monitoring (IHM). A well-established quantitative technique is Aircraft Oil Analysis (AOA), in which spectroscopic techniques such as Rotating Disk Electrode Atomic Emission Spectroscopy (RDE-AES) are employed to analyse periodic oil samples for wear debris. Usually, no sample preparation is undertaken, as the oil sample containing both dissolved and suspended metallic wear debris is analysed directly. AOA works well for oil-lubricated systems with relatively coarse filtration that allow circulation of the debris and its subsequent abrasive contact with moving components. To avoid this secondary wear, finer filtration is employed on both new and older aircraft. Less wear debris, and thus information, is available in the oil. A technique that quantitatively analyses the wear debris caught on the filter has been developed and is termed Quantitative Filter Debris Analysis (QFDA). Actual oil filters from CF88 Challenger ALF 502L-2C and Hornet F404 engines were obtained in sequence, when possible, prepared by the developed procedure, and analysed with AOA instrumentation. With sufficient results, both normal and abnormal levels of wear rates emerged, as has been recorded and used for AOA. Moreover, trending of the data for sequential samples has demonstrated the capability of GFDA for condition monitoring.     

Diesel engine fuel injection monitoring using acoustic measurements and independent component analysis

Air-borne acoustic based condition monitoring is a promising technique because of its intrusive nature and the rich information contained within the acoustic signals including all sources. However, the back ground noise contamination, interferences and the number of Internal Combustion Engine ICE vibro-acoustic sources preclude the extraction of condition information using this technique. Therefore, lower energy events; such as fuel injection, are buried within higher energy events and/or corrupted by background noise.     

Condition monitoring of electric motor ball bearings for the detection of grease contaminants

Grease used in the ball bearings of electric motors often get contaminated either from external particles or particles generated within these bearings. The effectiveness of vibration, stator current, acoustic emission and shock pulse measurements in detecting the presence of contaminant particles in bearing grease has been investigated. Silica and ferric oxide particles were used to contaminate grease. The levels of vibration, stator current, acoustic emission and shock pulse appreciably increased as contaminant level and contaminant size increased. Acoustic emission peak amplitude proved to be the best condition monitoring technique for the detection of grease contaminants in motor bearings. It is followed by shock pulse maximum value and carpet value in terms of effectiveness.     

Condition monitoring of planetary gearbox by hardware implementation of artificial neural networks

The hazards of planetary gearboxes’ failures are the most crucial in the machinery which directly influence human safety like aircrafts. But also in an industry their damages can cause the large economic losses. Planetary gearboxes are used in wind turbines which operate in non-stationary conditions and are exposed to extreme events. Also bucket-wheel excavators are equipped with high-power gearboxes that are exposed to shocks. Continuous monitoring of their condition is crucial in view of early failures, and to ensure safety of exploitation. Artificial neural networks allow for a quick and effective association of the symptoms with the condition of the machine. Extensive research shows that neural networks can be successfully used to recognize gearboxes’ failures; they allow for detection of new failures which were not known at the time of training and can be applied for identification of failures in variable-speed applications. In a majority of the studies conducted so far neural networks were implemented in the software, but for dedicated engineering applications the hardware implementation is being used increasingly, due to high efficiency, flexibility and resistant to harsh environmental conditions. In this paper, a hardware implementation of an artificial neural network designed for condition monitoring of a planetary gearbox is presented. The implementation was done on a Field Programmable Gate Array (FPGA). It is characterized by much higher efficiency and stability than the software one. To assess condition of a gearbox working in non-stationary conditions and for chosen failure modes, a signal pre-processing algorithm based on filtration and estimation of statistics from the vibration signal was used. Additionally, the rewards-punishments training process was improved for a selected neural network, which is based on a Learning Vector Quantization (LVQ) algorithm. Presented classifier can be used as an independent diagnostic system or can be combined with traditional data acquisition systems using FPGAs.     

Statistical process monitoring based on orthogonal multi-manifold projections and a novel variable contribution analysis

Multivariate statistical methods have been widely applied to develop data-based process monitoring models. Recently, a multi-manifold projections (MMP) algorithm was proposed for modeling and monitoring chemical industrial processes, the MMP is an effective tool for preserving the global and local geometric structure of the original data space in the reduced feature subspace, but it does not provide orthogonal basis functions for data reconstruction. Recognition of this issue, an improved version of MMP algorithm named orthogonal MMP (OMMP) is formulated. Based on the OMMP model, a further processing step and a different monitoring index are proposed to model and monitor the variation in the residual subspace. Additionally, a novel variable contribution analysis is presented for fault diagnosis by integrating the nearest in-control neighbor calculation and reconstruction-based contribution analysis. The validity and superiority of the proposed fault detection and diagnosis strategy are then validated through case studies on the Tennessee Eastman benchmark process.     

The influence of chromatic components on extended depth of field imaging

The analysis of variance is one of the most commonly used algorithms for detecting focalized zones in digital images and is an entry point to extended focalization techniques beyond those established by optical laws. This article analyses the dependence of the red, green, and blue (RGB) and lightness, hue, and saturation (LHS) components when used as the basis for applying an algorithm to obtain images with extended depth of field. Also, an algorithm developed by the authors is described and the dependence of the final result is shown according to the chromatic components used as the variables. Finally, a methodology is defined based on the study of second variance in relation to the number of images and pixels of the chromatic coordinates to decide which to use as the basis for the calculation. Microsc. Res. Tech., 2009. © 2009 Wiley-Liss, Inc.     

Simulation and analysis of vibration signals generated by rolling element bearing with defects

Dynamic loading of a rolling element bearing structure is modeled by a computer program developed in Visual Basic programming language. The vibration response of the structure to the dynamic loading is obtained using a standard finite element package I-DEAS. A force model is proposed to model the localized rolling element bearing defects. Time and frequency domain analyses are performed for diagnostics of rolling element bearing structures. Statistical properties of the vibration signals for healthy and defected structures are compared. The envelope (HFRT) method is employed in the frequency domain analysis. The effect of the rotational speed on the diagnostics of rolling element bearing defects is investigated. An optimum sensor location on the structure is sought. Effect of the structure geometry on the monitoring techniques is studied. An optimum monitoring method can be employed by analyzing the rolling element bearing structure following the procedure proposed in this study. The present commercial computer aided engineering packages can be used in special engineering applications such as condition monitoring of rolling element bearings.     

Self-Learning Algorithm for Automated Design of Condition Monitoring Systems for Milling Operations

This paper investigates an approach, termed self-learning ASPS (automated sensor and signal processing selection), aimed at aiding the systematic design of condition monitoring systems for machining operations. The paper outlines a self-learning methodology for the classification of the system’s normal and faulty states and the selection of the most appropriate sensors and signal processing methods for detecting machining faults in end milling. The aim of the proposed approach is to enable the condition monitoring designer to use previous system faults or incidents to design an on-line monitoring system, reducing the system’s development time and cost. Force, acceleration and acoustic emission signals are used to design the condition monitoring systems for end milling operations. Gradual tool wear, catastrophic cutter breakage and tool collision are used for evaluating the proposed self-learning ASPS approach. The initial results show that the suggested algorithm can be applied for an automated, self-learning monitoring system for the selection of the most sensitive sensors and signal processing methods for machining faults and conditions.       

动态压差控制阀流场及流量控制特性研究

针对变流量加热及冷却系统水力和热力失调的问题,设计一种动态压差控制阀.基于计算流体力学(CFD)方法,建立不同阀芯开度下动态压差控制阀三维流道模型.对比研究了不同阀芯开度下阀内流场分布以及流量变化,得出了动态压差控制阀在不同阀芯开度下阀内压降曲线的变化规律、阀芯节流口处速度曲线及湍动能曲线的分布规律,拟合了阀门出口流量...     

Characteristics analysis and experiment verification of electrostatic sensor for aero-engine exhaust gas monitoring

An electrostatic sensor is a critical component of the electrostatic monitoring system for an aero-engine gas path. In this paper, the basic principles of the electrostatic sensor are described and the materials selection process for sensor electrodes and an isolation medium is introduced. The finite element method was adopted to calculate the sensitivity distribution, and the influence of relevant structural parameters on the sensitivity distribution characteristic is analyzed. The data fitting method was employed to acquire the unified spatial sensitivity distribution functions for a given structure sensor, which provides a useful reference for the sensor’s installation location. Based on the unified function, the sensitivity distribution function along the particle moving direction and the frequency characteristic of the electrostatic sensor were acquired. Then the corresponding influence factors of the frequency properties were analyzed. An experiment platform was established to verify the model and the theoretical analysis results. Then, simulated experiments were applied to verify the feasibility and validity of the electrostatic sensor, and the experiment results provided a useful reference for the identification of abnormal particles as a characteristic of aero-engine faults.     

Condition monitoring of industrial machinery through lubricant oil analysis

A number of techniques exist that can be used to determine the condition of a machine by analysing a sample of the lubricant. However, there are only a few available techniques that can be used to determine the effect of more than one cause of deterioration with a single method of analysis. By sensing changes in the dielectric of the lubricant, it is possible to obtain information relating to both the deterioration of the machine by means of wear products, and the influence of any solid and liquid contaminants present in the lubricant. In this paper, the principle of operation for this method is described, and the results of some controlled tests are reported in which comparisons are made with results obtained using other types of ‘health’ monitor.     

Bivariate analysis of complex vibration data: An application to condition monitoring of rotating machinery

The problem of the robust definition of the acceptance regions in condition monitoring of the vibrations of rotating machinery is related to the more wide field of the analysis of bivariate data. Traditional parametric techniques and innovative nonparametric methods based on the statistical concept of the data depth are presented and critically examined in the paper. The performances with respect to the robustness in the estimation of the acceptance regions are analysed by means of experimental cases of real rotating machinery of a power plant.     

基于动态核独立元统计量的石油管道泄漏检测

石油管道泄漏是受腐蚀、磨损、焊缝缺陷、振动、冲刷以及人为破坏等多种因素影响的连续动态过程,单纯基于压力信号的检测和基于高斯分布假设的信号分析方法不能适应其多变量、强耦合、动态特性。为此,综合考虑与管道泄漏有关的操作参数和环境参数,针对管道监测参数呈现时序自相关性、泄漏检测精度不高的问题,提出一种基于动态核独立分量分析(DKICA)的石油管道泄漏检测方法。首先引入动态特性确定算法(DOD)计算模型最佳参数阶次,解决动态过程导致的监测参数呈现时序自相关性问题;再采用核独立分量分析(KICA)在核主元空间提取独立元;最后通过考察独立元的T2、SPE联合指标判断泄漏发生。通过对某一输送场站采集的数据进行实验验证,结果表明采用联合指标D2的正常样本误检率和泄漏样本漏检率都远低于单独采用T2或SPE统计量;而引入动态特性的2阶DKICA对于正常样本的误检率和泄漏样本的漏检率都低于未引入动态特性的KICA方法。可见,所提出的基于动态核独立分量联合指标的石油管道泄漏检测方法是一种高效且可行的方法。     

基于依赖性分析的企业级组件集成框架

论述了企业级组件集成框架的设计目标、设计思想和实现方案,运用可扩展标记语言描述企业级组件间的依赖关系,构建了一个能实现企业级组件即插即用的集成框架。     

石英挠性加速度计摆片组件的应力分析

针对石英挠性加速度计摆片组件的梁易产生断裂的现象,通过分析石英挠性加速度计的工作原理,利用Pro/E分别对石英摆片和铝合金骨架进行了三维建模并装配,然后根据载荷计算在ANSYS中对装配件进行了系统的应力分析,给出了石英摆片组件在实际偏移0.02 mm最大位移下的变形场和应力场,并根据应力场分布提出了石英摆片梁结构的优化方案。研究结果表明,石英摆片组件的最大应力随着加速度载荷的增加保持不变,且小于石英材料的许用应力。     

The diagnostic line: A novel criterion for condition monitoring of rotating machinery

This study examined scaling properties of an increment series from rotating machinery. Moreover, two fluctuation parameters for the smallest and largest time scales of a scaling range served as a pair of fluctuation parameters to describe system conditions. Therefore, an interesting phenomenon is observed: the data points, each representing a pair of fluctuation parameters, for fault conditions almost form a straight line, while those for normal clearly depart from the straight line. To describe the phenomenon, a novel concept termed the diagnostic line was introduced. Subsequently, properties of the diagnostic line were carefully investigated theoretically and numerically. Consequently, a decisive role of noise in forming the diagnostic line was determined. Accordingly, this study develops a novel criterion for condition monitoring of rotating machinery.