The present invention provides an accurate machine monitoring technique based on vibration analysis. An AR parametric model is generated to characterize a normal machine condition. Subsequently, data is collected from a machine during operation. This data is fit to the AR parametric model, and an Exponentially Weighted Moving Average (EWMA) statistic is derived therefrom. The EWMA statistic is able to identify whether the machine is in a normal state ("in control") or in an abnormal state ("out of control"). Additionally, an EWMA control chart is generated that distinguishes between normal and abnormal conditions, and between different abnormal conditions. As a result, once the EWMA statistic is generated, it is compared to the EWMA chart for determination of the specific fault that is ailing the machine.

Machine performance monitoring and fault classification using an exponentially weighted moving average scheme

A DSP-based measurement system dedicated to the vibration analysis on rotating machines was designed and realized. Vibration signals are on-line acquired and processed to obtain a continuous monitoring of the machine status. In case of a fault, the system is capable of isolating the fault with a high reliability. The paper describes in detail the approach followed to built up fault and non-fault models together with the chosen hardware and software solutions. A number of tests carried out on small-size three-phase asynchronous motors highlight the excellent promptness in detecting faults, low false alarm rate, and very good diagnostic performance.

A DSP-based FFT-analyzer for the fault diagnosis of rotating machine based on vibration analysis

A DSP-based measurement system dedicated to the vibration analysis on rotating machines was designed and realized. Vibration signals are on-line acquired and processed to obtain a continuous monitoring of the machine status. In case of fault, the system is capable of isolating the fault with a high reliability. The paper describes in detail the approach followed to built up fault and unfault models together with the chosen hardware and software solutions. A number of tests carried out on small-size three-phase asynchronous motors highlights high promptness in detecting faults, low false alarm rate, and very good diagnostic performance.

A digital signal processor (DSP)-based FFT analyzer capable of adapting its behavior according to specific application requirements was set up. A suitably designed user-interface allows both the specific application to be chosen and its main parameters to be defined. An automatic optimization procedure then reorganizes the instrument configuration in order to meet user requirements. Multi-DSP architecture implementation ensures real-time behavior. The experimental test results indicate that the instrument is capable of optimally operating in a large number of typical applications.

A multi-application FFT analyzer based on a DSP architecture

An intelligent FFT analyzer capable of adapting its operating parameters on the basis of the signal spectrum was set up and characterized. The realized instrument is based on a parameter optimization procedure which provides the instrument with auto-configuration capability. It was implemented on a multiple processor DSP architecture in order to achieve real-time behavior. The experimental tests carried out on a large number of signals highlight the instrument capability of correctly detecting tones with a good frequency resolution whatever the signal spectrum type.

An intelligent FFT-analyzer

The use of digital signal processors (DSPs) that enable continuous estimates of signal frequency components to be incorporated into a broad range of real-time systems is considered. Classical spectral estimation methods are reviewed. Real-time implementation of the methods, all of which use some form of fast Fourier transform method, on DSP chips is discussed. Some applications of these chips are examined. >

Y. Jain

Papers

Embedding spectral analysis in equipment