Time Domain Vibration Research Articles

Time-domain simulation and nonlinear analysis on ride performance of four-wheel vehicles

A nonlinear dynamic model with eight DOFs of a four-wheel vehicle is established in this paper. After detaching the nonlinear characteristics of the leaf springs and shock absorbers, the multi-step linearizing method is used to simulate the vehicle vibration in time domain, under a correlated four-wheel road roughness model. Experimental verifications suggest that the newly built vehicle model and simulation procedure are reasonable and feasible to be used in vehicle vibration analysis. Furthermore, some nonlinear factors of the leaf springs and shock absorbers, which affect the vehicle ride performance (or comfort), are investigated under different vehicle running speeds. Some substaintial rules of the nonlinear vehicle vibrations are revealed in this paper.

Application of the Generalised Likelihood Ratio Algorithm to the Detection of a Bearing Fault in a Helicopter Transmission

A Bell 206B main rotor gearbox was run at high load under test conditions in the Helicopter Transmission Test Facility operated by the Defence Science and Technology Organisation (DSTO) of Australia. The test succeeded in initiating and propagating pitting damage in one of the planet gear support bearings. Vibration acceleration signals were recorded periodically for the duration of the test. The time domain vibration signals were converted to angular domain to minimise the effects of speed variations. Auto-Regressive Moving-Average (ARMA) models were fitted to the vibration data and a change detection problem was formulated in terms of the Generalised Likelihood Ratio (GLR) algorithm. Two different forms of the GLR algorithm in window-limited online form were applied. Both methods succeeded in detecting a change in the vibration signals towards the end of the test. A companion paper submitted by the University of New South Wales outlines the corresponding diagnosis and prognosis algorithms applied to the vibration data.

시간영역에서 과도 비틀림 진동에 의한 저속 2행정 디젤엔진의 축계 피로강도 평가

Two stroke low speed diesel engines are widely used for marine propulsion or as power plant prime mover. These engines have many merits which includes higher thermal efficiency, mobility and durability. Yet various annoying vibrations occur sometimes in ships or at the plant itself. Of these vibrations, torsional vibration is very important and dictates a careful investigation during the engme`s initial design stage for safe operation. With the rule and limit on torsional vibration in place, shaft strength fatigue due to torsional vibration however demands further analysis which possibly can be incorporated in the classification societies` rule and limit. In addition, the shaft`s torsional vibration stresses can be calculated equivalently from accumulated fatigue cycles number due to transient torsional vibration in time domain. In this paper, authors suggest a new estimation method combined with Palmgren-Miner equation. A 6S70MC-C () engine for ship propulsion was selected as a case study. Angular velocity was measured, instead of shaft`s strain, for simplified measurement and it was converted to torsional vibration stress for accumulated fatigue cycle numbers in shafting life time. Likewise, the accumulated fatigue calculation was compared with shaft fatigue strength limit. This new method can be further realized and confirmed in ship with two stroke low speed diesel engine.

Fault Classification of Water Hydraulic System by Vibration Analysis with Support Vector Machine

Abstract This paper presents a new neural network approach to the fault diagnosis of a water hydraulic system based on the wavelet analysis of a vibration signal. A novel feature of this approach is that the vibration signals acquired from the water hydraulic motor are employed for analysis. Wavelet transform (WT) is first applied as a feature extraction technique to analyze the time-domain vibration signal. The performance of support vector machine (svm) is then investigated and compared with the conventional neural network. The results confirm the applicability of the proposed method for the fault detection in a modern water hydraulic system.

On the cross correlation function amplitude vector and its application to structural damage detection

A new approach to detecting structure damage using the cross correlation function amplitude vector (CorV) of the measured vibration responses is proposed. It is verified that under a steady random excitation with specific frequency spectrum, the CorV of a structure only depends on the frequency response function matrix of the structure, and it is also found that the normalized CorV has a specific shape. Thus the damage can be detected and located with the correlation and the relative difference between the CorVs obtained from intact and damaged structures. The Cross Correlation Function Amplitude Vector Assurance Criterion (CVAC) is then defined and can be used to quantify the variation of CorV. It is found that the CVAC decreases monotonously with the increasing of damage factor, which indicates the change of CorV is related to the damage severity. The methods of damage locating with CorV are then proposed and demonstrated by the experiments. Finally, the experiment on detection in the fasteners loosing of an aircraft panel model are presented to illustrate the application of the CorV. The feature of this approach lies in that the CorV is obtained from the time domain vibration responses of the structure under steady random excitation, and it has the advantages of simplicity in calculation and the damage detection, so it is possible that the presented approach applies to the structural health monitoring (SHM) with steady ambient excitations.

Expert system development for vibration analysis in machine condition monitoring

Expert systems can be adapted for machine condition monitoring data interpretation due to the ability to identify systematic reasoning processes. As vibration analysis in condition monitoring is still generally performed by highly trained professionals, the use of expert systems would allow a greater analysis throughput as well as enabling technicians to perform routine analysis. The development of an expert system for vibration analysis of fixed plant is discussed, as well as laboratory and industry testing. Unique to existing developments, the expert system incorporates triaxial and demodulated frequency and time domain vibration data analysis algorithms for high accuracy fault detection. The tests confirm the potential value of the expert system for both laboratory and on-site maintenance departments of large manufacturing and mineral processing plants.

Comparison of integrated micro-electrical-mechanical system and piezoelectric accelerometers for machine condition monitoring

The design and implementation of instrumentation to collect real-time vibrational data from a quasi-steady state machine (a dry vacuum pump) for fault prediction diagnostics is presented. When simultaneous multiple data collection points are required on the same machinery, the use of conventional transducers such as piezoelectric accelerometers becomes impractical due to their price, as each needs an expensive associated charge amplifier. The use of inexpensive micromachined integrated micro-electrical-mechanical system accelerometers such as ADXL105 has been explored here as an alternative to piezoelectric accelerometers for obtaining reliable and predictable data for diagnostics. Surface micromachined accelerometers are a new technology and their usage for vibrational analysis has been conservative due to concerns over increased noise levels and tolerance to high temperatures. In this article, it is shown that such concerns can be allayed. The time and frequency domain vibration signatures obtained using both types of accelerometers are compared. The study shows that ADXL105 accelerometers can be an effective alternative low-cost high-quality solution for machine condition monitoring.

Artificial neural networks and genetic algorithm for bearing fault detection

A study is presented to compare the performance of three types of artificial neural network (ANN), namely, multi layer perceptron (MLP), radial basis function (RBF) network and probabilistic neural network (PNN), for bearing fault detection. Features are extracted from time domain vibration signals, without and with preprocessing, of a rotating machine with normal and defective bearings. The extracted features are used as inputs to all three ANN classifiers: MLP, RBF and PNN for two- class (normal or fault) recognition. Genetic algorithms (GAs) have been used to select the characteristic parameters of the classifiers and the input features. For each trial, the ANNs are trained with a subset of the experimental data for known machine conditions. The ANNs are tested using the remaining set of data. The procedure is illustrated using the experimental vibration data of a rotating machine. The roles of different vibration signals and preprocessing techniques are investigated. The results show the effectiveness of the features and the classifiers in detection of machine condition.

1409 回転機械の異常振動診断システムの開発 : 転がり軸受の異常診断(GS-10 回転機,研究発表講演)

This paper describes the technique for detecting an abnormal vibration of the rolling bearing of a practical rotating machinery. The rolling bearing is diagnosed from time domain vibration data measured by an accelerometer, and spectrum data obtained by the frequency analysis. Two diagnostic methods are proposed. One is the case that vibration data can be obtained from the normal rolling bearing. The other is the case that vibration data cannot be obtained from the normal rolling bearing. Then, the effectiveness of the diagnosis system was verified for the rolling bearing with defective ball, outer race and inner race.

Cyclic time averaging for machine monitoring

A cyclic time averaging method is directed at processing time-domain vibration data of the machine to produce an averaged time waveform over a selected cyclic time period. It is not necessary to employ a triggering signal to obtain the averaged time waveform.

Bearing Fault Detection Using Artificial Neural Networks and Genetic Algorithm

A study is presented to compare the performance of bearing fault detection using three types of artificial neural networks (ANNs), namely, multilayer perceptron (MLP), radial basis function (RBF) network, and probabilistic neural network (PNN). The time domain vibration signals of a rotating machine with normal and defective bearings are processed for feature extraction. The extracted features from original and preprocessed signals are used as inputs to all three ANN classifiers: MLP, RBF, and PNN for two-class (normal or fault) recognition. The characteristic parameters like number of nodes in the hidden layer of MLP and the width of RBF, in case of RBF and PNN along with the selection of input features, are optimized using genetic algorithms (GA). For each trial, the ANNs are trained with a subset of the experimental data for known machine conditions. The ANNs are tested using the remaining set of data. The procedure is illustrated using the experimental vibration data of a rotating machine with and without bearing faults. The results show the relative effectiveness of three classifiers in detection of the bearing condition.

Artificial neural networks and support vector machines with genetic algorithm for bearing fault detection

A study is presented to compare the performance of bearing fault detection using two different classifiers, namely, artificial neural networks (ANNs) and support vector machines (SMVs). The time-domain vibration signals of a rotating machine with normal and defective bearings are processed for feature extraction. The extracted features from original and preprocessed signals are used as inputs to the classifiers for two-class (normal or fault) recognition. The classifier parameters, e.g., the number of nodes in the hidden layer in case of ANNs and the radial basis function kernel parameter (width) in case of SVMs along with the selection of input features are optimized using genetic algorithms. The classifiers are trained with a subset of the experimental data for known machine conditions and are tested using the remaining set of data. The procedure is illustrated using the experimental vibration data of a rotating machine. The roles of different vibration signals and signal preprocessing techniques are investigated. The results show the effectiveness of the features and the classifiers in detection of machine condition.

Gear fault detection using artificial neural networks and support vector machines with genetic algorithms

A study is presented to compare the performance of gear fault detection using artificial neural networks (ANNs) and support vector machines (SMVs). The time-domain vibration signals of a rotating machine with normal and defective gears are processed for feature extraction. The extracted features from original and preprocessed signals are used as inputs to both classifiers based on ANNs and SVMs for two-class (normal or fault) recognition. The number of nodes in the hidden layer, in case of ANNs, and the radial basis function kernel parameter, in case of SVMs, along with the selection of input features are optimised using genetic algorithms (GAs). For each trial, the ANNs and SVMs are trained with a subset of the experimental data for known machine conditions. The trained ANNs and SVMs are tested using the remaining set of data. The procedure is illustrated using the experimental vibration data of a gearbox. The roles of different vibration signals, obtained under both normal and light loads, and at low and high sampling rates, are investigated. The results compare the effectiveness of both types of classifiers without and with GA-based selection of features and the classifier parameters. For most of the cases considered, the classification accuracy of SVM is better than ANN, without GA. With GA-based selection, the performance of both classifiers are comparable, in most cases, with three selected features. However, for SVMs with six features, 100% classification success is achieved in all test cases. The training time of SVMs is substantially less compared to ANNs in all cases considered. The present classification accuracy compares well with the results reported in a recent work, (Mech. Systems Signal Process. 16 (2002) 373), though the data and the feature sets are different.

ARTIFICIAL NEURAL NETWORK BASED FAULT DIAGNOSTICS OF ROLLING ELEMENT BEARINGS USING TIME-DOMAIN FEATURES

A procedure is presented for fault diagnosis of rolling element bearings through artificial neural network (ANN). The characteristic features of time-domain vibration signals of the rotating machinery with normal and defective bearings have been used as inputs to the ANN consisting of input, hidden and output layers. The features are obtained from direct processing of the signal segments using very simple preprocessing. The input layer consists of five nodes, one each for root mean square, variance, skewness, kurtosis and normalised sixth central moment of the time-domain vibration signals. The inputs are normalised in the range of 0.0 and 1.0 except for the skewness which is normalised between −1.0 and 1.0. The output layer consists of two binary nodes indicating the status of the machine—normal or defective bearings. Two hidden layers with different number of neurons have been used. The ANN is trained using backpropagation algorithm with a subset of the experimental data for known machine conditions. The ANN is tested using the remaining set of data. The effects of some preprocessing techniques like high-pass, band-pass filtration, envelope detection (demodulation) and wavelet transform of the vibration signals, prior to feature extraction, are also studied. The results show the effectiveness of the ANN in diagnosis of the machine condition. The proposed procedure requires only a few features extracted from the measured vibration data either directly or with simple preprocessing. The reduced number of inputs leads to faster training requiring far less iterations making the procedure suitable for on-line condition monitoring and diagnostics of machines.

Predicting Soil and Structure Vibrations from Impact Machines

A method for predicting the complete time-domain vibration records of soil and structures prior to installation of foundations for impact machines is described in detail. A paper provides support for the validity of accepted assumptions about the use of impacts directly on the soil for estimating the impulse response functions, and the effect of load zone dimensions on soil vibrations. The procedures for experimental testing on a site and computing the predicted complete vibration records are developed. Resulting patterns of computed and measured vibrations are shown for block-type foundation and a foundation under a vibroisolated block for impact machine. Furthermore, the method enables us to consider the heterogeneity and variety of soil and structure properties. In particular, it is useful under nonuniform and complicated soil conditions for determination and verification of safe distances from machine foundations to objects sensitive to vibrations.

Vibration arthrometry in the patients with failed total knee replacement

This is a preliminary research on the vibration arthrometry of artificial knee joint in vivo. Analyzing the vibration signals measured from the accelerometer on patella, there are two speed protocols in knee kinematics: 1) 2 degrees/s, the signal is called "physiological patellofemoral crepitus (PPC)", and 2) 67 degrees/s, the signal is called "vibration signal in rapid knee motion". The study has collected 14 patients who had revision total knee arthroplasty due to prosthetic wear or malalignment represent the failed total knee replacement (FTKR), and 12 patients who had just undergone the primary total knee arthroplasty in the past two to six months and have currently no knee pain represent the normal total knee replacement (NTKR). FTKR is clinically divided into three categories: metal wear, polyethylene wear of the patellar component, and no wear but with prosthesis malalignment. In PPC, the value of root mean square (rms) is used as a parameter; in vibration signals in rapid knee motion, autoregressive modeling is used for adaptive segmentation and extracting the dominant pole of each signal segment to calculate the spectral power ratios in f < 100 Hz and f > 500 Hz. It was found that in the case of metal wear, the rms value of PPC signal is far greater than a knee joint with polyethylene wear and without wear, i.e., PPC signal appears only in metal wear. As for vibration signals in rapid knee motion, prominent time-domain vibration signals could be found in the FTKR patients with either polyethylene or metal wear of the patellar component. We also found that for normal knee joint, the spectral power ratio of dominant poles has nearly 80% distribution in f < 100 Hz, is between 50% and 70% for knee with polyethylene wear and below 30% for metal wear, whereas in f > 500 Hz, spectral power ratio of dominant poles has over 30% distribution in metal wear but only nonsignificant distribution in polyethylene wear, no wear, and normal knee. The results show that vibration signals in rapid knee motion can be used for effectively detecting polyethylene wear of the patellar component in the early stage, while PPC signals can only be used to detect prosthetic metal wear in the late stage.

Statistical tool breakage detection schemes based on vibration signals in NC milling

We develop a vibration sensor-based tool breakage detection system for NC milling operations. The system obtains the time-domain vibration signal from the sensor attached on the spindle bracket of our CNC machine and declares tool failures through the on-line monitoring schemes. For the on-line detection, our approach is to use the statistical process control methods where control limits or thresholds are automatically calculated independently of cutting conditions. The main thrust of this paper is to compare the performance of the proposed statistical process monitoring methods including the X-bar control scheme, the exponentially weighted moving average (EWMA) scheme, and the adaptive EWMA scheme. The performance of the control schemes are compared in terms of the type I and II errors calculated from the experiment data.

ARTIFICIAL NEURAL NETWORK BASED FAULT DIAGNOSTICS OF ROTATING MACHINERY USING WAVELET TRANSFORMS AS A PREPROCESSOR

The purpose of condition monitoring and fault diagnostics are to detect and distinguish faults occurring in machinery, in order to provide a significant improvement in plant economy, reduce operational and maintenance costs and improve the level of safety. The condition of a model drive-line, consisting of various interconnected rotating parts, including an actual vehicle gearbox, two bearing housings, and an electric motor, all connected via flexible couplings and loaded by a disc brake, was investigated. This model drive-line was run in its normal condition, and then single and multiple faults were introduced intentionally to the gearbox, and to the one of the bearing housings. These single and multiple faults studied on the drive-line were typical bearing and gear faults which may develop during normal and continuous operation of this kind of rotating machinery. This paper presents the investigation carried out in order to study both bearing and gear faults introduced first separately as a single fault and then together as multiple faults to the drive-line. The real time domain vibration signals obtained for the drive-line were preprocessed by wavelet transforms for the neural network to perform fault detection and identify the exact kinds of fault occurring in the model drive-line. It is shown that by using multilayer artificial neural networks on the sets of preprocessed data by wavelet transforms, single and multiple faults were successfully detected and classified into distinct groups.

Wavelet frame expansions and wavelet response functions

Wavelet frame expansions are utilized to introduce ‘‘wavelet response functions’’ (WRF). The convolution integral is replaced by an efficient synthesis scheme by incorporating multiresolution and wavelet frame expansion concepts. Wavelet frame expansion is used to partition a wide spectral range into multiresolution frequency bands. The response from each resolution level is obtained by using a sampling interval that is matched with the bandwidth. The contributions coming from different resolution levels are combined to construct the overall system response. In the applications, the attention is focused on the synthesis of the transient time-domain vibration response of proportionally damped (Bn=η, fn, n=modes) wideband linear systems. Naturally, such systems, having impulse response durations that are inversely proportional to the modal bandwidths (Tmaxn= 2.2/Bn), require a multiresolution synthesis scheme during the construction of their transient response, which involves fast-decaying high-frequency components that are superimposed on top of slowly decaying low-frequency modal responses. The proposed construction technique averts the time-domain aliasing problem associated with the DFT schemes. The effectiveness of the wavelet frame synthesis scheme is demonstrated by considering the construction of the transient vibration response of a finite dispersive system.

Determination Of Rigid Body Characteristics From Time Domain Modal Test Data

A new method is presented to identify the characteristics of a rigid body and its supports, such as the center of gravity, the moment of inertia, and the stiffness and damping matrices, based on the measured time domain vibration data. Transformation matrices are formed using only the geometric co-ordinates relative to an arbitrarily selected origin to relate the pure translational measurement to both translation and rotation of the rigid body system, The rigid body modes and mode shapes, carefully selected from the accurately determined modes and mode shapes (rigid body and structural), are used, together with the special matrix structure, such as symmetry. The test result shows that the center of gravity and the moment of inertia are determined with good accuracy. The simple test procedure makes it an easily implementable method in practice.