Showing papers in "Mechanical Systems and Signal Processing in 1992"

TL;DR: In this article, an effective procedure for vibration condition monitoring of low speed (⩽100 RPM) rolling element bearings is described, which incorporates fault detection using frequency domain trending indices with fault diagnosis using the frequency spectra.

TL;DR: In this article, peak amplitude in the frequency domain, peak RMS, and the power spectrum were used as indirect indices to develop a system for monitoring and classifying ball bearing defects, which was then processed by artificial neural networks.

TL;DR: In this article, a stochastic model structure that explicitly accounts for both the input-output and noise dynamics is proposed for non-linear Wiener-Hammerstein-type systems.

TL;DR: In this article, an effective procedure for automatic fault diagnosis in low speed (⩽100 RPM) rolling element bearings is described, which involves the calculation of a statistical distance measure between vibration signals.

TL;DR: In this article, an indirect method for detecting a transient source waveform is introduced, where a vibration response signal that includes information on an excitation pulse and dispersive structural characteristics is monitored by using a sensor at a remote position, and cepstral analysis plays the role of a robust inverse filter to smooth out the transmission path and deverberate the complicated vibration response.

TL;DR: In this paper, a smoothing method consisting of estimating by least-squares the coefficients of a two-dimensional discrete Fourier series with arbitrary period and arbitrary frequency resolution is presented, which can be used with non-equally-spaced and non-rectangular-domain data.

TL;DR: In this article, a new method for in-process detection of chatter in milling, which is based on the probability distribution of the cutting force signal, is proposed, and investigated using computer simulations and verified experimentally under various cutting conditions.

TL;DR: In this paper, the authors developed a parameter estimation algorithm for the stepped sine identification of the transfer function of a linear system when all measurements are corrupted by noise, which consists of three steps: (1) an overdetermined set of linear equations containing stepped-sine measurements is solved in a linear least squares sense.

TL;DR: In this paper, the authors developed the concept of selective sensitivity for detection of stiffness abnormalities in beams by static displacement measurement and demonstrated analytically for a discrete elastic model of a beam that a system of applied forces can always be chosen to produce maximum deflection sensitivity to change in selected stiffness constants, while at the same time causing zero sensitivity to changes of all other spring constants.

TL;DR: In this article, a method is presented in which an initial mass or modal mass estimate is refined by examining the sensitivity of identified parameters to the estimated mass using data from two or more excitation frequencies.

TL;DR: An analysis of four widely accepted non-parametric spectral estimation methods: Periodogram, Blackman-Tukey, Welch, and Nuttall-Carter is presented and it is found that all methods can achieve the same mean, variance, resolution, and resolution but differ in leakage suppression and computational complexity.

TL;DR: The orthogonal estimator which is introduced to curve fit to the complex frequency response functions provides information on the model structure and the unknown parameter values for linear and non-linear continuous time models.

TL;DR: In this paper, an experimental modal analysis technique applied to small scale models to identify tubular joint damage which very often occurs in offshore framed structures was used to determine the cause of the tubular damage.

TL;DR: In this article, the effect of windings, frame, and impregnation on the vibrational behavior of stators of electrical machines is investigated. But the analysis is based on the three-dimensional theory of elasticity and is limited to stators with encased construction.

TL;DR: Algorithms that can be used to estimate statistical moments of both Gaussian and beta distributions for vibration signals applications are presented, demonstrating here that fitting the raw data on the beta distribution reduces the sensitivity to noise and thus guards against false alarms.

TL;DR: It is shown that the inverse filtering process is equivalent to that of prediction error estimation based on a signal model in the form of an autoregressive moving-average (ARMA) model.

TL;DR: The approach, which employs gradient adaptive lattice analysis and pattern recognition techniques, is fast and yields accurate recognition of tool wear states in a wide range of cutting conditions.

TL;DR: In this article, the use of the backwards prediction error model for modal parameter identification is examined, and the philosophies behind techniques based on forwards and backwards models are compared, while the Kumaresan and Tufts approach to determine system modes from spurious modes by the orientation of the estimated eigenvalues on the complex plane is applied to both simulated and real data.

TL;DR: A method for determination of priority among inputs when applying the partial coherence function technique to source identification problems and is applied to a simulated two-inputs/one-output system in which the inputs were correlated to each other.

TL;DR: In this article, the average power intensity compressed in each equivalent sequency group of the Haar transformed data is extracted as the characteristic parameter named the impulse index which enables high-speed real-time processing and fast on-line fault detection.

TL;DR: In this article, the authors discussed the errors that are introduced when calculating the impulse response function from the frequency response function using DFT/FFT analysis technique and showed that the estimated impulse response functions are not described by a single exponential decaying sinewave but rather as a sum of exponential decaying sewaves, and applied this slightly more complex model when the measurements are to be used for mathematical modelling of systems containing lightly damped resonances.

TL;DR: A unified recursive form of the damped complex exponential method is presented, encompassing the features of some well known time-domain identification algorithms, which permits the minimisation of data acquisition and time consumption, but also provides more flexible techniques for problems such as time varying modal parameters and structural control.

TL;DR: In this article, a non-linear vehicle model with tunable shock absorbers is presented for random road excitations, and a generalised discrete harmonic linearization technique based upon a processed energy function is developed to characterise nonlinear restoring and damping forces by an array of equivalent linear stiffness coefficients.

TL;DR: In this paper, two extensions to the coherence function for isolating components of power-spectral density measurements are derived and tested, and the use of the latter is illustrated in the removal of contaminants from measurements of wall-pressure spectra and bundles of cylinders in turbulent axial flow.

TL;DR: In this article, a combined path and force control system is realized by putting most control efforts into a feed forward loop with off-line properties and only minor control activities into the on-line loops.

TL;DR: In this article, two real-time solutions for time-delay estimation problems using adaptive digital filters are evaluated. And they are implemented on a Texas Instruments TMS320-C25 system and detailed performance comparisons are given.

TL;DR: In this paper, a methodology based on recovery of the acceleration pattern of the final actuator of the mechanism under consideration seems to be particularly promising, which allows the working conditions to be judged, directly comparing the recovered output member acceleration with the functional requirements which, for this kind of mechanism, are mainly relative to the motion law.

TL;DR: In this paper, the selection of the sampling interval for estimating the modal parameters from a randomly excited structure is considered, and the optimum design of uniform data sampling intervals when experimental constraints allow only a limited number of discrete-time measurements of the output from the continuous system is discussed.

TL;DR: In this paper, the authors discuss the feasibility of an experimental method to detect non-linear structural behavior in a practical engineering environment based on the linear analysis of nonlinear systems with increasing amplitude random excitation signals.

TL;DR: In this article, the authors investigated the different techniques offered to the analyst in modal testing for the estimation of the number of modes present in the response of a structure or system under test.