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.
It is currently prevalent to locate faults for a satellite power system based on an expert system, not utilizing all the available information provided by tests. The casual network model for a satellite power system is presented. Considerations for failure probability of each component of the power system, the cost of applying each test, the influence of a precedent test result on the next test selection, and an optimal sequential testing algorithm for fault location is presented. This program is applied to locate the failure component of the power system of a satellite. The results show this program is very effective and it is very fast to generate an optimal diagnosis tree.
卫星诊断软件的组件化平台技术是当前开放系统结构(Open System Architecture for Condition-Based Maintenances,OSA-CBM)研究的一项应用技术。组件技术保证了软件的通用性,平台技术提供了软件的开放性。结合我国卫星诊断软件的现状,阐述了组件平台技术的基本理论,给出了基于组件平台技术的卫星诊断软件开发过程。最后通过一个开发实例,说明了组件平台技术的实用性。
