Dynamic error separation is still an important subject and research high tide nowadays. Along with the development of time series and modern spectrum analysis, dynamic error separation combined with Mathematical Modeling becomes a main trend. A new algorithm for separating dynamic error is proposed in this paper. It is based on time series and modern spectrum modeling principle. Firstly, dynamic random error is separated by dint of time series model combining recursive instrumental variables and Kalman filter. Then the trending component and the harmonic component are respectively identified by using Stepwise Rejection-Accept Regression Analysis and Pisarenko spectrum analysis method. Finally, compared with the prior information, dynamic system error is mostly separated. Separation of the error components is verified through simulation experiment. The performances of algorithm are illustrated according to the results obtained from simulation. It shows that 99.53% of random errors are separated, in addition, the estimation of trend component and frequencies of periodic components (the corresponding amplitudes) of system error are also obtained.

Date Published: 31 December 2008
PDF: 6 pages
Proc. SPIE 7130, Fourth International Symposium on Precision Mechanical Measurements, 71305M (31 December 2008); doi: 10.1117/12.819762

Published in SPIE Proceedings Vol. 7130:
Fourth International Symposium on Precision Mechanical Measurements
Yetai Fei; Kuang-Chao Fan; Rongsheng Lu, Editor(s)