论文信息 - Sequential tests for failure detection

Sequential tests for failure detection

Abstract The problem of constructing sequential tests for fault detection methods that are based on the concept of Mahalanobis distance is addressed. First the distribution of the Mahalanobis distance that belongs to the degraded (alternative) state of a multivariate system is determined. Then three sequential tests are constructed and their performance investigated for fault detection. For each test, the average sample number and an upper limit for the number of samples necessary that the test will terminate with a given probability are calculated.

Imre Pázsit | I. Pázsit

[1] Julius T. Tou,et al. Pattern Recognition Principles , 1974 .

[2] A. Erdélyi,et al. Tables of integral transforms , 1955 .

[3] Martin Adams,et al. A sequential failure detection technique and its application , 1976 .

[4] J. Edward Jackson,et al. Sequential x 2 - and T2-Tests and Their Application to an Acceptance Sampling Problem , 1961 .

[5] A. Erdélyi,et al. Higher Transcendental Functions , 1954 .

[6] J. Edward Jackson,et al. Sequential $\chi^2$- and $T^2$-Tests , 1961 .

[7] G. Barrie Wetherill,et al. Sequential methods in statistics , 1967 .

[8] Athanasios Papoulis,et al. Probability, Random Variables and Stochastic Processes , 1965 .

[9] R. C. Kryter,et al. A Multivariate Statistical Pattern Recognition System for Reactor Noise Analysis , 1976, IEEE Transactions on Nuclear Science.