Condition Monitoring Model Research Articles

Spectral kurtosis based on AR model for fault diagnosis and condition monitoring of rolling bearing

Spectral kurtosis (SK) is an algorithm that gives an indication of how kurtosis varies with frequency. A frequency band that contains abundant information, especially the impact signal, can be tracked by calculating SK. In the present article, SK combined with Autoregressive AR model, was applied into the fault diagnosis and condition monitoring of bearings. Accelerated life test of rolling bearings in Hangzhou Bearing Test & Research Center (HBRC) was performed to collect vibration data over their entire lifetime (normal-fault-failure). The result shows that SK can detect early incipient fault by eliminating some other interfering frequency components. In addition, it can detect fault 5 min earlier than root mean value (RMS). This fault detection in advance is significant for condition monitoring.

Genetic algorithm-based wrapper approach for grouping condition monitoring signals of nuclear power plant components

Equipment condition monitoring of nuclear power plants requires to optimally group the usually very large number of signals and to develop for each identified group a separate condition monitoring model. In this paper we propose an approach to optima

Condition Monitoring of CI Engine Running on Biodiesel Using Transient Process

Various research groups across the globe have developed models for engine condition monitoring and fault diagnosis based on the data from steady state performance measurements over the last two decade. However, these performance data are not amenable for easy collection in some situations and some of the data may show little performance deviation at the steady state conditions of operation. The aim of this study is to identify engine operating conditions and parameters that can be used to develop a diagnostic tool for internal combustion (CI) engine running on biodiesel blends fuel. Ricardo Wave Software was used to stimulate healthy and degraded engine. The degraded condition of operation was created by forcing one injector to work with 90%, 80% and 70% of the normal fuel flow efficiency. The measurable performance parameters such as brake power, brake torque and exhaust temperature and CO emission have been used to compare the deviations during steady and transient condition. From the comparison of the data obtained from transient and steady state simulation of the engine with and without injector fault, it can be concluded that the transient parameters show higher deviations and hence are better suited for condition monitoring and diagnostic modelling in engine working speed.

Development of a tool wear observer model for online tool condition monitoring and control in machining nickel-based alloys

Online monitoring and in-process control improves machining quality and efficiency in the drive towards intelligent machining. It is particularly significant in machining difficult-to-machine materials like super alloys. This paper attempts to develop a tool wear observer model for flank wear monitoring in machining nickel-based alloys. The model can be implemented in an online tool wear monitoring system which predicts the actual state of tool wear in real time by measuring the cutting force variations. The correlation between the cutting force components and the flank wear width has been established through experimental studies. It was used in an observer model, which uses control theory to reconstruct the flank wear development from the cutting force signal obtained through online measurements. The monitoring method can be implemented as an outer feedback control loop in an adaptive machining system.

Multisegment pulverised coal mill model and online implementation for condition monitoring

This paper presents a new multisegment coal mill mathematical model, which covers the whole milling process from mill start-up to shutdown. This model is initially derived through analysing energy transfer, heat exchange and mass flow balances of a coal mill. Evolutionary computational techniques are then adopted to identify the unknown system parameters. After the model was tested offline, it was converted into a discrete-time form for the purpose of online implementation. A shared-memory technique is employed to accommodate the measurement data dynamically in realtime, which provides a generic platform to communicate with different software environments used by different mill plants. A user-friendly interface is designed to satisfy the requirement specified by the plant engineers. The model can be used to improve the combustion process control and fault diagnosis.

State space models for condition monitoring: a case study

A Condition Monitoring system can increase safety, quality and availability in industrial plants. Safety requirements are especially important in critical machineries, like a turbine driving a centrifugal compressor located at a petrochemical plant in the case study presented in this paper. A Condition Monitoring system is set up for vibration data coming from the turbine. Four years of monthly data observed at two different locations of the equipment are analysed. The core of the system is a model to forecast the state of the machine using data provided by the Condition Monitoring system at each moment in time. The model is based on the State Space framework whose associated recursive algorithms (Kalman Filter and Fixed Interval Smoothing) provide the basis for a number of different operations, from which the most important in the present context is the extrapolation of the distribution of forecasts on which the probability of failure is estimated. The cost model on which the decision of making a preventive replacement is taken is based on the ‘expected cost per unit time’ for a pre-determined critical value of the vibration measure. The system is thoroughly tested on the data.

A Hybrid Model for Tool Condition Monitoring and Optimal Tool Management

This paper introduces a new hybrid model for tool condition monitoring (TCM) and optimal tool management (OTM) in end milling operation. The model includes a wavelet fuzzy neural network with acoustic emission (AE) and a model of fuzzy classification of tool wear state with the detected cutting parameters supported by cutting database. The results estimated by cutting conditions and detected signals are fused by artificial neural network (ANN) so as to facilitate effective tool replacement at a proper state or time. The validity and reliability of the method are verified by experimental results.

Estimation of parameters in a linear state space model using a Rao-Blackwellised particle filter

A Rao-Blackwellised particle filter is used in the estimation of the parameters of a linear stochastic state space model. The proposed method combines the particle filtering technique with a Kalman filter using marginalisation so as to make full use of the analytically tractable structure of the model. Simulation studies are performed on a simple illustrative example and the results demonstrate the effectiveness of the proposed method in comparison with the conventional extended-Kalman-filler-based method. The proposed method is then applied in the estimation of the parameters in a railway vehicle dynamic model for condition monitoring and the desired results have been obtained.

Study of fuzzy neural networks model for system condition monitoring of AUV

A structure equivalent model of fuzzy-neural networks for system condition monitoring is proposed, whose outputs are the condition or the degree of fault occurring in some parts of the system. This network is composed of six layers of neurons, which represent the membership functions, fuzzy rules and outputs respectively. The structure parameters and weights are obtained by processing off-line learning, and the fuzzy rules are derived from the experience. The results of the computer simulation for the autonomous underwater vehicle condition monitoring based on this fuzzy-neural networks show that the network is efficient and feasible in gaining the condition information or the degree of fault of the two main propellers.

Evolving wavelet networks for power transformer condition monitoring

This paper proposes a novel model for power transformer condition monitoring using evolving wavelet networks (EWNs). The EWNs are three-layer structures, which contain wavelet, weighting and summing layers. The EWNs automatically adjust the network parameters, translation, and dilation in the wavelet nodes and the weighting values in the weighting nodes, through an evolutionary based optimization process. Global search abilities of the evolutionary algorithm as well as the multiresolution and localization natures of the wavelets enable the EWNs to identify the complicated, numerical-knowledge relations of dissolved gas contents in transformer oil to corresponding fault types. The proposed EWNs have been tested on the Taipower Company diagnostic records and compared with the fuzzy diagnosis system, artificial neural networks as well as the conventional method. The test results reveal that the EWNs possess far superior diagnosis accuracy and require less constructing time than the existing methods.

Towards Practical and Synthetical Modelling of Repairable Systems

In this article, we survey the developments with respect to generalized models of repairable systems during the 1990s, particularly for the last five years. In this field, we notice the sharp fundamental problem that voluminous and complicated models are proposed without sufficient evidence (or data) for justifying a success in tackling real engineering problems. Instead of following the myth of using simple models to face complicated reality, and based on our own research experiences, we select and review some practical models, in the quickly growing areas: age models, condition monitoring models, and shock and wear models, including the delay-time models. Further, we also notice that there is an attempt to develop synthetical models from a different point of view. Therefore, we comment the relevant developments with strong emphasis on stochastic processes reflecting the intrinsic nature of the actual physical dynamics of those repairable system models.

A Dynamic Model for Condition Monitoring of a High-Power CO2 Industrial Laser

Industrial laser systems handle high power consumptions and may function under undesirable operating conditions if the systems are not properly maintained. It is sometimes difficult to diagnose why a laser is not functioning properly because the optical output is the result of complex interactions among many parameters such as the total gas pressure, effectiveness of the laser cooling system, operating environment, and gradual deterioration of laser components. In this paper, a dynamic power distribution model is developed to characterize the power distribution of a high-power transverse-flow DC-excited CO2 laser to account for dynamic effects such as continuously ramping up and down the laser output power and the cyclic nature of the chiller. The model contains the essential dynamic features of a CO2 laser system and yields solutions sufficiently accurate for practical diagnostic purposes.

A state space condition monitoring model for furnace erosion prediction and replacement

The paper develops a replacement action decision aid for a key furnace component subject to condition monitoring. A state space model is used to predict the erosion condition of the inductors in an induction furnace in which a measure of the conductance ratio (CR) is used to indirectly assess the relative condition of the inductors, and to guide replacement decisions. This study seeks to improve on this decision process by establishing the relationship between CR and the erosion condition of the inductors. To establish such a relationship, a state space model has been established and the system parameters estimated from CR data. A replacement cost model to balance at any time costly replacements with possible catastrophic failure is also proposed based upon the predicted probability of inductor erosion conditional upon all available information. The well known Kalman filter is employed to derive the predicted and updated probability of inductor erosion level conditional upon CR data to date. This is the first time the condition monitoring decision process has been modelled for real plant based upon filtering theory. The model fits the data well, gives a sensible answer to the actual problem, and is transferable to other condition monitoring contexts. Possible extensions are discussed in the paper.

A simple condition monitoring model for a direct monitoring process

This paper addresses the problem of condition monitoring of a component which has available a measure of condition called wear. Wear accumulates over time and monitoring inspections are performed at chosen times to monitor and measure the cumulative wear. If past measurements of wear are available up to the present, and the component is still active, the decision problem is to choose an appropriate time for the next inspection based upon the condition information obtained to date. A simple model which minimizes the expected cost per unit time over the time interval between the current inspection and the next inspection time is derived, and numerical examples are given to demonstrate the solution method.