Showing papers on "Condition monitoring published in 1991"
14 Apr 1991
TL;DR: A hidden Markov model (HMM)-based approach to mechanical system monitoring is presented and it is shown to be useful for machining applications with the associated problems of tool wear detection and prediction.
Abstract: A hidden Markov model (HMM)-based approach to mechanical system monitoring is presented. The resulting system is shown to be useful for machining applications with the associated problems of tool wear detection and prediction. The approach is based on continuous density, left-right HMMs that closely match the one-way, fresh-to-worn transition process of machining tools. The Baum-Welch iterative training procedure is modified to incorporate prior knowledge of the transitions between tool wear states. Results presented demonstrate that a multisensor HMM-based system is an effective approach for tool wear detection and prediction. >
63 citations
TL;DR: Delays time analysis is used to model the maintenance practices for coal face machinery within British Coal to predict the effectiveness of condition monitoring using the resultant downtime of machinery as the relevant measure.
27 citations
20 citations
TL;DR: A suite of computer programs has been developed which enables maintenance personnel to access, display and analyze in-flight event data recorded by the IECMS and to track the performance of individual engines based on "health indices" derived from the takeoff ground roll recordings as mentioned in this paper.
Abstract: The F404-GE-400 engines of the CF-18 aircraft are the first engines of the Canadian Forces to be maintained under a formal on-condition maintenance program. In support of this program, the Canadian Forces are developing advanced troubleshooting and performance monitoring procedures based on information recorded by the aircraft In-flight Engine Condition Monitoring System (IECMS). A suite of computer programs has been developed which enables maintenance personnel to access, display and analyze in-flight event data recorded by the IECMS and to track the performance of individual engines based on “health indices” derived from the IECMS takeoff ground roll recordings. The new techniques have been under evaluation at each of the CF-18 main operating bases for a period of approximately 14 months. Results to-date indicate that the IECMS recordings provide a considerable amount of information of benefit to engine technicians and maintenance planners.Copyright © 1990 by ASME
18 citations
TL;DR: The role of monitoring instruments is studied, and the results show that the monitoring of a system not only can reduce, but also can increase the entropy of the system.
Abstract: The complexity of a system arises not only due to the number of components and the characteristics of each component, but also to the types of interconnections between the components. A method of measuring this structural complexity is suggested as reflected in system operational difficulty using an information theoretic approach. This method, the functional entropy measure, can denote the structural complexity induced from the interconnections between components. An equation for calculating the diagnostic entropy without introducing any approximation is derived. The role of monitoring instruments is also studied, and the results show that the monitoring of a system not only can reduce, but also can increase the entropy of the system. To know the true condition of a monitored system with far greater confidence than if it remains unmonitored, the monitoring instrument must be much more reliable than the monitored component. >
15 citations
01 Jan 1991
TL;DR: The requirements for increased plant productivity and safety, and reduced maintenance costs, have led to an increasing interest in methods for condition monitoring of mechanical systems as mentioned in this paper, and this interest has led to the development of a number of methods for monitoring mechanical systems.
Abstract: The requirements for increased plant productivity and safety, and reduced maintenance costs, have led to an increasing interest in methods for condition monitoring of mechanical systems. In the li ...
15 citations
14 May 1991
TL;DR: A description is given of how the unconditioned signal from the sensor in a turbine flowmeter can be analyzed to provide a signature of the instrument.
Abstract: How the unconditioned signal from the sensor in a turbine flowmeter can be analyzed to provide a signature of the instrument is described. The signature indicates the condition of the instrument within the process environment and it adds to the customary process measurement information provided by a turbine flowmeter. A theoretical model is used to synthesize the basic signal expected from a turbine flowmeter based on an assumption that the instrument was operated under normal and abnormal conditions. The results obtained by analyzing synthesized and measured signals are compared and used to validate the theoretical model. The ideas presented have wide-ranging ramifications for both users and manufacturers of turbine flowmeters. >
14 citations
08 Jul 1991
TL;DR: In this article, the authors describe new insulation diagnostic and monitoring techniques, developed with the Queensland Electricity Commission, for use in an in-service HV plant, which enable the dielectric dissipation factor (DDF) and partial discharge (PD) level of HV insulation to be measured while an instrument transformer or bushing is in service.
Abstract: Insulation condition monitoring of high-voltage instrument transformers and transformer bushings is of significant importance in maintaining a safe and reliable electricity supply system. To achieve this, there is an increasing trend to perform in-situ insulation quality measurements on this equipment. The authors describe new insulation diagnostic and monitoring techniques, developed with the Queensland Electricity Commission, for use in an in-service HV plant. The techniques enable the dielectric dissipation factor (DDF) and partial discharge (PD) level of HV insulation to be measured while an instrument transformer or bushing is in-service. Application of the DDF measurement technique to provide a continuous insulation monitor is also covered. The continuous online system is designed to ultimately allow monitoring of all bushings and current transformers within an entire situation. >
14 citations
TL;DR: It was shown that the Pareto distribution can be used and the Paretian symptom reliability curve can be transformed to the average symptom life-curve of a given machine set.
13 citations
TL;DR: In this article, a mathematical model has been developed to predict the root-mean-square (RMS) value of the acoustic emission signal in milling, which is directly proportional to the power expended in turning.
Abstract: The monitoring of tool wear is important in maintaining the quality of workpieces produced. This paper presents a methodology to monitor on-line tool wear in end milling using acoustic emission. It is well known that the root-mean-square (RMS) value of the acoustic emission is directly proportional to the power expended in turning. A mathematical model has been developed to predict the RMS value of the acoustic-emission signal in milling. This mathematical model incorporates the machining parameters as variables. The accuracy of the model has been verified by a series of experiments. The experiments were carried out on a Bridgeport milling machine and data was collected and analysed by using an on-line computer data acquisition system. A comparison of the experimental and theoretical RMS values indicates a very good agreement between them. A control strategy similar to the moving average/moving range charts has been developed for monitoring on-line tool wear. The limits for the control charts were obtaine...
12 citations
14 May 1991
TL;DR: In this article, a parametric modeling technique has been applied to the predictive maintenance of rotating machinery, where the residual signal obtained by computing the difference between the predicted and measured signal is used in the prediction procedure.
Abstract: A parametric modeling technique has been applied to the predictive maintenance of rotating machinery. The case, which is explained in detail, was taken from an experiment in which a one-step gearbox was run at about 150% of nominal load until failure occurred. During the experiment, vibration signals from the gearbox were measured. The novelty of this work is to analyze the residual signal obtained by computing the difference between the predicted and measured signal. A parametric modeling technique called autoregressive modeling is utilized in the prediction procedure. The analysis of the residual signal proved to be successful; the failure could be predicted easier and earlier than using traditional fast Fourier transform (FFT)-based methods. >
TL;DR: The condition monitoring techniques are overviewed and some examples of current longwall mining machinery applications are presented and an expert system for condition monitoring purposes is discussed together with several case histories for underground mining machinery maintenance programs.
01 Nov 1991
TL;DR: In this paper, the fundamental equation of vibration and the disturbing forces involved is considered and the concept of L10 life is introduced and some of the methods for predicting the onset of failure are described.
Abstract: In any industrial fan, the rotating elements are located within some form of bearing arrangement. Defects, such as unbalance, will lead to an increase in vibration which is transmitted via these elements to the fan supports. As the bearings wear or as impeller balance deteriorates due to the effects of erosion, corrosion or dust build-up, there will be a corresponding increase in vibration magnitude. The rate of change of amplitude is a useful indication of the fan's ‘health’. The paper considers the fundamental equation of vibration and the disturbing forces involved. It is shown that a baseline fan signature is necessary if maximum benefits are to be achieved by any fan health programme. Ultimately, well-designed and installed bearings will fail due to fatigue. The concept of L10 life is also introduced and some of the methods for predicting the onset of failure are described.
18 Nov 1991
TL;DR: Reports a set of experiments performed to test the ability of a backpropagation network to classify the condition of an operating desktop fan based on its vibration signature.
Abstract: Reports a set of experiments performed to test the ability of a backpropagation network to classify the condition of an operating desktop fan based on its vibration signature The trained network was used to detect and classify faults commonly occurring in industrial fans, ie impeller unbalance and cracked impeller blade The discussion of the experimental results raises a number of issues relating to the application of this technique to the condition monitoring of industrial fans >
21 Oct 1991
TL;DR: A three-level process monitoring and diagnostics system has been developed for a plasma etching cell that incorporates generic cell controller concepts and some of the SEMI Generic Equipment Model (GEM) specifications.
Abstract: A three-level process monitoring and diagnostics system has been developed for a plasma etching cell. The physical and logical layout of this system is designed to co-exist with a proposed hierarchical cell control system. The cell control system design incorporates generic cell controller concepts and some of the SEMI Generic Equipment Model (GEM) specifications. At the lowest level of the system, equipment and process parameter monitoring is achieved. An AT-compatible computer serves as an information collection/router agent as well as an interpreter (gateway) between the SECS/RS-232 formatted messages received from the equipment controller system and the higher-level expert systems commands. At the top of the hierarchy, an expert systems package that resides on a SUN SPARC station and operates on an X-Windows environment collects information on the equipment and process from the AT compatible. The many system capabilities that result include: (1) pseudo real-time equipment and process monitoring, (2) equipment diagnostics and alarm reporting, (3) alarm logging; and (4) real-time equipment control. >
01 Sep 1991
TL;DR: In this article, a modified LMS (least mean square) algorithm is proposed in order to model the impulse-like signals with non-stationarity, which is suitable to real-time monitoring during the process of working operation and fault diagnostics of reciprocating machines.
Abstract: The spectrum analysis by means of FFT (fast Fourier transform) analysis has been widely used in mechanical signature analysis. An alternative method is proposed in the paper for condition monitoring and diagnostics of reciprocating machines. A modified LMS (least mean square) algorithm is proposed in order to model the impulse-like signals with non-stationarity. Convergence of the algorithm is investigated. The approach presented is suitable to real-time monitoring during the process of working operation and fault diagnostics of reciprocating machines.
01 Jan 1991
01 Jan 1991
Proceedings Article•
11 Sep 1991
TL;DR: In this paper, the authors describe a very simple method of detecting a slight rotor asymmetry which generally signifies the beginning of rotor failures which, if left unattended, could lead to a much more serious stator damage or machine breakdown.
Abstract: Over the years the manfacturers of induction motors have tried their best to produce reliable induction machines for the motor industry A good monitoring system which could detect rotor cage faults which may arise during manufacturing or ealy signs of impending motor failure in the rotor would therefore serve a very useful purpose The author describes a very simple method of detecting a slight rotor asymmetry which generally signifies the beginning of rotor failures which, if left unattended, could lead to a much more serious stator damage or machine breakdownn
Journal Article•
TL;DR: In this paper, the results of an investigation of flaw condition monitoring using ultrasonic arrays of an optimised design, directly coupled to solid, conductive test subjects are presented, showing that accurate, quantitative flaw characterisation is possible, and that the array structure has attractive features which should permit its application using several different monitoring strategies.
Abstract: Condition monitoring of structures, rather than the use of a series of isolated inspections, is becoming of greater interest as the capabilities of ultrasonic NDT systems increase. In this paper, some results of an investigation of flaw condition monitoring utilising ultrasonic arrays of an optimised design, directly coupled to solid, conductive test subjects are presented. It is shown that accurate, quantitative flaw characterisation is possible, and that the array structure has attractive features which should permit its application using several different monitoring strategies.
TL;DR: Neural network based adaptive methods can significantly improve the symptom interpretation in case of malfunctions of mathematically difficult to describe systems and processes.
12 Feb 1991
TL;DR: A team of researchers at the Robert Gordon Institute of Technology (RGIT) in Aberdeen have researched, developed, and applied new online condition monitoring techniques and diagnostic instrumentation systems for three-phase electrical machines.
Abstract: A team of researchers at the Robert Gordon Institute of Technology (RGIT) in Aberdeen have researched, developed, and applied new online condition monitoring techniques and diagnostic instrumentation systems for three-phase electrical machines. The author provides an overview of this research work through to its industrial application. Reference is made to the fundamental studies of fault mechanisms and signal processing techniques required to extract the information from signals which are characteristic of a malfunction in an electrical machine. The philosophy of approach to the research and development work is mentioned and the importance of carrying out extensive field trials prior to the commercial development of stand-alone instrumentation systems is emphasised. Laboratory and on-site results are presented and the new range of condition monitoring projects being researched at RGIT are mentioned.
26 Oct 1991
TL;DR: This - paper reviews the use of screening, and statistical process control Statistical Product Monitoring (SPM) of significant component electrical parameters, to techniques, permits the development of develop a component quality strategy.
Abstract: As the quality of integrated circuits has improved Statistical Product Monitoring, an over the last several years to 100 DPPM or less, it is no longer practical-for high volume users of low innovative test strategy, based on variables DPPM devices to rely on direct product testing to data collection, environmental stress control the quality of devices going into their products. This - paper reviews the use of screening, and statistical process control Statistical Product Monitoring (SPM) of significant component electrical parameters, to techniques, permits the development of develop a component quality strategy. The role of improved quality and reliability programs. While existing quality and reliability SPM in supplier qualification, selection, and continuous supplier monitoring is discussed. Practical examples from an on-line SPM system programs are used to determine the initial will be presented.
TL;DR: In this article, the authors outline the present status of plant condition monitoring activities at Scunthorpe works and provide an insight into the most important techniques used and the results achieved, focusing on the performance improvements and financial savings generated by systematic use of the following major condition monitoring techniques: thermography, electro-optical laser, vibration (portable data collection system).
Abstract: This paper aims to outline the present status of plant condition monitoring activities at Scunthorpe works and to provide an insight into the most important techniques used and the results achieved. Condition monitoring must produce positive results in terms of plant reliability, availability and safety and these must be quantifiable economically.This paper concentrates on the performance improvements and financial savings generated by systematic use of the following major condition monitoring techniques : thermography, electro-optical laser, vibration (portable data-collection system).
Journal Article•
19 Jun 1991
TL;DR: A monitoring system is presented, which detects errors in the execution of sensory robot programs and employs sensor models, in which both the measurement precision and the probability of misinterpretation are taken into account.
Abstract: A monitoring system is presented, which detects errors in the execution of sensory robot programs. To this aim the system plans and executes sensing strategies. It employs sensor models, in which both the measurement precision, and the probability of misinterpretation are taken into account. >
Dissertation•
01 Jan 1991
TL;DR: Advanced condition monitoring (ACM) technologies developed for in situ turbomachinery applications are reviewed in this article, based on direct in situ hardware monitoring and between-flight inspections, using novel real-time, automated, noncontacting, and nonintrusive sensor and associated electronic technologies.
Abstract: Advanced condition monitoring (ACM) technologies developed for in situ turbomachinery applications are reviewed. The ACM concepts are based on direct in situ hardware monitoring and between-flight inspections, using novel real-time, automated, noncontacting, and nonintrusive sensor and associated electronic technologies.