This paper investigates the characteristics of a non-stationary time series, which exists in mechanical fault diagnosis. Combining the characteristics with predictive efficiency, the limitation of the ARIMA model pred...This paper investigates the characteristics of a non-stationary time series, which exists in mechanical fault diagnosis. Combining the characteristics with predictive efficiency, the limitation of the ARIMA model prediction method is analyzed. This model often is applied in the prediction of a non-stationary times series in present. Thus, a wavelet prediction method is introduced to solve non-stationary problems. The Mallat method, often used in signal processing, results form the decimation or the retention of one out of every two samples. Its advantage is that just enough information is kept to allow the exact reconstruction of the input series, but the disadvantage is a time-varying series on line cannot be pursued. Therefore, the authors present another method, à Trous method, which can be applied for recursive prediction in real-time sampling procedure.展开更多
Today there is a big interest in reducing the maintenance costs and in increasing the reliability of machines in continuous operation. Therefore, maintenance on condition is used. State-of-the-art is a trend analysis ...Today there is a big interest in reducing the maintenance costs and in increasing the reliability of machines in continuous operation. Therefore, maintenance on condition is used. State-of-the-art is a trend analysis and a fault prediction made only based on sensor signals and stochastic methods. The identification possibilities of this technique are limited. A new concept for model-based monitoring has been developed for more detailed fault identification. The developed concept determines the condition of a machine after the occurrence of a fault. The concept is based on a simulation including various faults and an optimization tool. The development of a cost function and the optimization is one of the challenges of such a concept. Using an AMB rotor system with an auxiliary bearing, the new concept of model-based monitoring is investigated using experiments and the optimization is discussed in this paper.展开更多
基金Sponsored by the National High Technology Research and Development Program of China (Grant No.2002AA721063).
文摘This paper investigates the characteristics of a non-stationary time series, which exists in mechanical fault diagnosis. Combining the characteristics with predictive efficiency, the limitation of the ARIMA model prediction method is analyzed. This model often is applied in the prediction of a non-stationary times series in present. Thus, a wavelet prediction method is introduced to solve non-stationary problems. The Mallat method, often used in signal processing, results form the decimation or the retention of one out of every two samples. Its advantage is that just enough information is kept to allow the exact reconstruction of the input series, but the disadvantage is a time-varying series on line cannot be pursued. Therefore, the authors present another method, à Trous method, which can be applied for recursive prediction in real-time sampling procedure.
基金supported by a fellowship within the Postdoc-Programme of the German Academic Exchange Service (DAAD)
文摘Today there is a big interest in reducing the maintenance costs and in increasing the reliability of machines in continuous operation. Therefore, maintenance on condition is used. State-of-the-art is a trend analysis and a fault prediction made only based on sensor signals and stochastic methods. The identification possibilities of this technique are limited. A new concept for model-based monitoring has been developed for more detailed fault identification. The developed concept determines the condition of a machine after the occurrence of a fault. The concept is based on a simulation including various faults and an optimization tool. The development of a cost function and the optimization is one of the challenges of such a concept. Using an AMB rotor system with an auxiliary bearing, the new concept of model-based monitoring is investigated using experiments and the optimization is discussed in this paper.
