Detection Of Incipient Failures Research Articles

Detection of incipient failures by using an [formula omitted]-norm criterion: Application to railway switching points

The paper deals with condition monitoring of electric switch machines for railway points. The proposed detection system is based on off-line processing of armature current and voltage sampled as the machine operates a switch of railway tracks. These data are normally available from conventional railway signalling infrastructures; no additional transducers and instrumentation are required. The system, which basically consists of an algorithm tuned on a model of the machine behaviour as both the rails are simultaneously driven towards their rest position, allows detection of faults of an incremental nature, specifically those connected to progressive increasing of frictional loads due to loss of lubrication, deterioration of slide chairs and increasing obstructions. The algorithm implements finite impulse response systems whose convolution profiles are designed on the basis of an H 2 -norm type criterion which guarantees robustness, particularly with respect to electrical noise.

Detection of Incipient Failures by using Geometric Approach Techniques: An Application to Railway Switching Points

The design of observer units for detecting anomalous operating conditions in linear dynamic systems is carried out, in the dual setting of noninteraction, on the basis of an H2-norm type criterion also including relevant feasibility constraints. The observer units are implemented as discrete-time FIR systems. A specific application is also considered, which consists in health monitoring of an electric switch machine for railway points.

Measurement of synchronous generator data from digital fault recorders for tracking of parameters and field degradation detection

This paper presents a description of advanced instrumentation technology developed at Arizona Public Service (APS) for on-line monitoring of large turbogenerators. These signals include three-phase generator voltages and currents, generator field winding voltages and currents, generator power angle and turbine front standard speed deviation. The data is collected by a 16 analog channel 10 bit digital fault recorder (DFR) for several operating conditions including steady state, transient events due to transmission line faults, etc., and staged tests. The paper also discusses the methodology behind the use of these measured signals for on-line tracking of the stability parameters of a synchronous generator. In this context, a procedure for the on-line estimation of machine parameters and turns ratios will be developed. Continuous monitoring of the machine parameters and turns ratios in real time may then be used for incipient failure detection.

Neural Network-Based Fault Detections in a Pressurized Water Reactor Pressurizer

The early detection of incipient failures is of paramount importance for the safety and reliability of nuclear power plants. The feasibility of using artificial neural networks as process simulators in a fault detection device is explored. Two neural networks are trained to follow the dynamic evolution of the system pressure in a nonfaulty pressurizer of a pressurized water reactor. During an accident, the discrepancy between the plant's signals and the neural networks' predictions can be used to rapidly detect the faulty condition. In reality, the signals will be unavoidably affected by a certain level of noise. The robustness of neural networks to noisy patterns assures a satisfactory degree of accuracy in the process predictions and, therefore, a high efficiency in the detection as well.

96/05003 Potential exposure in nuclear safety

In the frame of extensive work for the development of incipient failure detection methods, neutron noise signals in PWRs of different designs have been investigated using correlation techniques and interpreted by means of theoretical and experimental analysis of the vibrational behaviour of reactor internals. As driving forces of the dominant neutron noise sources have been identified: core barrel beam mode vibrations, shell mode vibrations of cylindrical components and fuel assembly beam mode vibrations. Variations in the long term behaviour of the ex-vessel neutron noise spectra are shown and related to changes in the mechanical properties of fuel assemblies.

Incipient Fault Detection in Rotating Machinery by Using Artificial Neural Networks

Detection of incipient failures is a fundamental issue in routing machines. A new approach for on-line detection of incipient faults is proposed. This methodology is based on vibration monitoring dau and fuzzy art neural networks. Numerical results demonstrating the high efficiency of the proposed approach are presented.

An acoustic diagnostic technique for use with electric machine insulation

Many of the electrical failures of large electric machines have mechanical precursors. This premise has led to the attempts described here to detect mechanical changes, such as groundwall delamination, by acoustic methods as a means of early detection of incipient failure. By optimally launching an ultrasonic wave into a stator bar and using the conductor as a waveguide, it has been shown possible to interrogate the groundwall insulation and the critical interface region adjacent to the conductor. Laboratory studies using an epoxy mica stator bar subjected to thermal cycling have demonstrated that the acoustic signatures obtained with this method are mirrored by conventional corona spectroscopy conducted simultaneously, although earlier damage is clearly discernible. Based on the results of this study, the acoustic technique appears to have substantial promise as an advanced detection tool. >

Acoustic diagnosis of the onset of strut fracture of the BS/CC prosthetic heart valve

A replacement prosthetic heart valve functions in a corrosive environment in which it cannot readily be inspected or repaired. It must cycle repeatedly, at a rate of 40 million cycles per year, over a service life that may extend many years. The opening and closing sounds of Bjoerk–Shiley convexo‐concave prosthetic heart valves are brief transients that are being studied for detection of incipient valve failure. The sounds are produced by the impact of the occluder disk with the metallic struts that retain it within an orifice ring. These sounds have been recorded in vivo for two distinct conditions or classes, valves that are intact, and valves that exhibit an in‐service failure of one of two legs of a strut which is critical to the function of the valve. The small size and rapid motion of the valve occluder disk produce a broadband acoustic transient that exhibits subtle changes as a valve strut proceeds toward failure. Features are detected, and a nonlinear expansion is used to develop a set of coeffic...

A systematic approach for detection of malfunctions in equipment of nuclear desalination plants during operation

Two advanced surveillance systems were developed for on-line or off-line malfunction monitoring of plant components and detection of incipient failures of various plant equipment, using a System Dynamics Synthesis (SDS) method. Utilizing regression time series analysis of fluctuations associated with observable and measurable plant data, collected from already installed sensors and instrumentation signals, the SDS method allows for detection of small deviations from normal operation. Such slight changes are unlikely to be reflected in the mean value of operation signals, displayed to the operator, although undetected abnormalities may lead to unanticipated component failure and unscheduled plant shutdowns. Although malfunction prediction techniques based on analysis of information embeded in signal fluctuations were developed to enhance the performance of nuclear-powered desalination plants, the techniques can be applied to components of fossil-powered desalination plants using RO, ED, or distillation technologies. Generally, data collected from signals of a desalination plant instrumentation carry invaluable information on equipment condition as well as on the level of interaction between various components.

Magnescope: a portable magnetic inspection system for evaluation of steel structures and components

The authors describe a portable magnetic inspection device developed for nondestructive evaluation of ferromagnetic steel structures and components. The device consists of an inspection head, instrumentation for measuring magnetic-field and flux density, and a computer together with the necessary control and data-analysis software. The device has been used for nondestructive evaluation of the mechanical condition of steels, particularly, detection of stress and monitoring the progress of fatigue damage. The most important application is in the detection of incipient failure, such as occurs with fatigue damage, temper embrittlement, and creep damage. Results which demonstrate the successful use of this instrument in monitoring fatigue are described.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Performance Assessment and Statistical Process Control: An Approach to Operation and Maintenance Cost Reduction

Accurate and timely Information on the performance and condition of equipment is vital to safe, reliable and economical operation of power plants. Early detection of performance degradation and incipient failure is critical to the process of averting catastrophic damage and unscheduled outages.The Utility Performance Assessment package developed by Bailey Controls Integrates performance assessment with digital distributed data acquisition and control systems to provide an overall Information management system to assist optimal control and maintenance management of plant operations. Furthermore, Statistical Process Control techniques utilizing Shew hart Charts can be readily combined with the performance assessment package to detect plant abnormalities. It is the purpose of this contribution to highlight the features of the performance assessment package and to demonstrate how Statistical Process Control techniques can be coupled with a performance assessment package to produce an effective online tool detect equipment and reduce plant operation and maintenance costs.

A vector space solution to incipient sensor failure detection with applications to automotive environments

An efficient method of sensor diagnosis in electronically controlled IC engines is demonstrated. The approach taken is based on the failure detection and isolation (FDI) theory, and the resulting diagnostic systems can be implemented in on-board digital computers of passenger vehicles. The proposed diagnostic system, referred to as the binary phase detection filter (BPDF), uses a pair of failure-sensitive filters. Based on the reference model of an internal combustion engine, BPDF not only isolates the failed sensor, but also estimates the associated amount of failure. A wide variety of sensor failures, including soft partial types (e.g. calibration or bias drift), can be detected using BPDF. >

A computer-based multipoint vibration system for process plant rotating equipment

The authors describe the installation of a computer-based multipoint bearing vibration monitoring system in an oil refinery's process plant. The multipoint system, installed on centrifugal pumps, fans, and compressors in the 40 to 450 hp range, was developed in response to several serious plant fires which originated at centrifugal pump sea/bearing failures. It is concluded that detection of incipient mechanical failures has met the original objectives to reduce fire risk, production losses, and machinery downtime. Despite the limitations of the single probe on most items of equipment and the relatively course alarm setting, the system predicts maintenance problems and thereby reduces costs. >

Advanced burst processing methods in loose parts monitoring

Acoustic monitoring for detection of loosened or detached parts is recognized as an important technique for incipient failure detection at primary system of nuclear power plants. Appropriate hardware systems are installed at all German plants. The precision of the analysis statements, however, is essentially dependent on the used diagnosis methods. Increased application of computer based interactive systems for the analysis of acoustic burst signals can enhance the security of results considerably. Thus specific analysis tools have been developed comprehending the following items: • - Verification of burst impact events in acoustic signals, • - time analysis of series of impact events, • - plant-specific source location, • - application of Hertz theory for determination of energy and mass of impacting part or component. The corresponding methods are now tentatively used in the frame of measurement analyses. Experiences with this methods and results will be presented.

Indian developments in ultrasonics and acoustic emission methods

Fundamental investigations in ultrasonics in India date back to the early 20th century. But, fundamental and applied research in the field of nondestructive evaluation (NDE) came much later. In the last four decades it has grown steadily in academic institutions, national laboratories and industry. Currently, commensurate with rapid industrial growth and realisation of the benefits of NDE, the activity is becoming much stronger, deeper, broader and very wide spread. Acoustic Emission (AE) is a recent entry into the field of nondestructive evaluation. Pioneering efforts in India in AE were carried out at the Indian Institute of Science in the early 1970s. The nuclear industry was the first to utilise it. Current activity in AE in the country spans materials research, incipient failure detection, integrity evaluation of structures, fracture mechanics studies and rock mechanics. In this paper, we attempt to project the current scenario in ultrasonics and acoustic emission research in India.

The influence of noise diagnostic techniques on the safety and availability of nuclear power plants

Diagnostic techniques based on vibration and noise analysis have been developed in FRG with the aim of incipient failure and malfunction detection in nuclear power plants. On the basis of the Vibration Monitoring System in the nuclear power station GKN, the measuring system, the diagnosis concept and the interpretation results will be discussed in more detail. In order to achieve full acceptance for these methods a completely understanding of signal sources, long-term trends and alert levels is necessary. Some newer results of successful interpretation of signature deviations are described. Discussion of the benefits shows that noise diagnostic techniques can increase the plant availability and improve the operational safety.

Simulation and analysis of machinery fault signals

Machinery condition monitoring is rapidly finding applications in all branches of industry. In particular, vibration monitoring is playing an increasingly important role as a tool for assisting with predictive and preventive maintenance and for improving operation efficiency of plant. Condition monitoring systems are used for the detection of incipient failure and the diagnosis of the nature of faults in operating machinery. However, for these systems to be reliable an improved understanding is required of the vibration signatures produced by machinery failure mechanisms and of methods for the interpretation of these signals. Many types of fault produce vibration signals which are impulsive in nature and which may be buried in background noise. A method is described for simulating this type of signal and modelling the various stages of incipient failure. Statistical and spectral analysis are used to describe the fault development. The influence of machinery frequency response characteristics on signal transmission from the damaged are to the measurement point are also considered.

Identification of Multivariate Models for Noise Analysis of Nuclear plant

During the normal operation of a pressurized water reactor, neutron noise analysis with multivariate autoregressive procedures is a valuable diagnostic tool to extract dynamic characteristics for incipient failure detection. The first part of the paper will describe in details the equations for estimating the multivariate autoregressive model matrices and the struc.ture of various matrices. The matrices are estimated by solving a set of matrix operations, called Yule-Walker equations. The selection of optimal model order will also be discussed. Once the optimal parameter set is obtained, simple and fast calculations are used to determine the auto power spectral density, cross spectra, coherence function, phase. In addition the spectra may be decomposed into components being contributed from different noise sources.An application using neutron flux data collected on a nuclear plant will illustrate the efficiency of the method.

The application of vibration signature analysis and acoustic emission source location to on-line condition monitoring of anti-friction bearings

The mechanical noise, vibration and heat generated by defects in fabrications, process plant and machinery are a measure of component condition, and once characterised allow on-line assessment of defect severity. Unexpected failures which might threaten the entire production process can thereby be avoided. There is an increasing need for reliable and robust condition monitoring instrumentation with on-line data processing, particularly as the trend contunues towards larger production units with little or no standby capacity for use in the event of unscheduled stoppages due to mechanical failure. This article describes the detection of incipient failure of rolling element bearings by kurtosis and the location of fatigue cracks in slowly rotating bearings by acoustic emission

Assessment of Adriamycin cardiotoxicity in children by systolic time intervals

In a prospective study, 44 children receiving Adriamycin (ADR) for various neoplastic diseases underwent serial estimations of the systolic time intervals (STI) for the noninvasive assessment of left ventricular myocardial performance. Five of the 44 children developed clinical signs of ADR-related congestive heart failure at a cumulative dose of less than 550 mg/m2 body surface area. Clinical symptoms, changes in the electrocardiogram and in the chest X-ray were preceded in every case by changes of the STI, mainly a prolongation of the left ventricular pre-ejection period (PEP), or a decrease of the ejection time (ET)/pre-ejection period (PEP) ratio (ET/PEP). A continous increase of the PEP and a decrease of the ET/PEP-ratio also gave an indication of myocardial dysfunction during ADR treatment in the other children without clinical signs of congestive heart failure. This subclinical cardiotoxic effect of ADR below the critical cumulative dose of 550 mg/m2 was observed in children with pre-existent myocardial damage, with preceding thoracic irradiation, or during concurrent chemotherapy, of which cyclophosphamide seemed to be most important. Thus, the estimation of the STI proved helpful and reliable in the early detection of incipient heart failure and in the selection of high risk patients in children receiving ADR treatment.