The degraded failure of on-board Doppler Velocity Sensor (DVS), which achieves non-contact velocity measurement based on Doppler principle, can be mainly attributed to the aging of microwave modules and deviation of the radar emission angle. For the microwave modules, active devices such as Gunn diodes are prior in degradation with respect to other passive devices, with the phase noise expanding monotonically. On the other hand, the emission angle of antenna deviates due to the metro vibration. In view of the actual working condition of metro, the DVS may also suffer external shocks during the natural degradation process, which is mixed with the natural degradation by model of compound Poisson process in this paper. In view of the non-reversibility of degradation, the inverse Gaussian process is chosen to describe the gradual degradation of DVS. In addition, given the inherent and postnatal differences among individual products, such as the dislocation of active devices induced during the thermos-compression bonding and individual installation error of antenna, the drift coefficients in the model are randomized. On this basis, the impact of external shock is introduced into the reliability analysis competing with the natural degradation of components. Finally, through parameters estimation of virtual degradation testing data by simulation, the methodology is demonstrated.
Degradation Process DVS Inverse Gaussian process Poisson shock Randomized coefficients
This is a preview of subscription content, log in to check access.
This paper was co-supported by the Natural Science Foundation of Beijing Municipality (L171003), National Natural Science Foundation of China (51620105010, 51575019), and Program 111 of China.
Meeker, W.Q., Escobar, L.A.: Statistical Methods for Reliability Data. Wiley, New York (1998)zbMATHGoogle Scholar
Nelson, W.: Accelerated Testing: Statistical Models, Test Plans, and Data Analysis. Wiley, New York (1990)CrossRefGoogle Scholar
Shi, Y.: Bayesian methods for accelerated destructive degradation test planning. IEEE Trans. Reliab. 61, 245–253 (2012)CrossRefGoogle Scholar
Zhang, Z.: A prognostic approach for systems subject to Wiener degradation process with cumulative-type random shocks. In: 6th Data Driven Control and Learning Systems Conference, pp. 694–698. IEEE, Chongqing, China (2017)Google Scholar
Peng, C.: Inverse Gaussian processes with random effects and explanatory variables for degradation data. Technometrics 57(1), 100–111 (2015)MathSciNetCrossRefGoogle Scholar