22 November 2011 A new noise suppression algorithm for optical fiber temperature surveillance of heavy oil thermal recovery well
Author Affiliations +
Proceedings Volume 8199, 2011 International Conference on Optical Instruments and Technology: Optical Sensors and Applications; 81991A (2011) https://doi.org/10.1117/12.906604
Event: International Conference on Optical Instruments and Technology (OIT2011), 2011, Beijing, Beijing, China
Abstract
Pure silica core optical fiber is commonly used as the sensing fiber in Raman-backscatter distributed temperature sensors (DTS) in heavy oil thermal well. However the sensing signal collected from this type of fiber statistically belongs to nonstationary random process which cannot be effectively de-noised by simply applying conventional methods. To solve this problem, we develop a novel noise suppression algorithm by combining wavelet multi-scale analysis and moving grey model GM(1,1). The algorithm first applies wavelet de-noising in spatial domain of temperature profile to remove the high frequency noise, then uses moving GM(1,1) method to remove both high frequency and low frequency nonstationary noise in time domain. Autoregressive (AR) model and least square regression are used to optimize the forecasting parameters of GM(1,1). Finally the results of both domains are reconstructed to obtain the de-noised profile. Long-term field test was proposed on the Karamay oil field F11051 steam stimulation well, Xinjiang Province, China. Field test result shows that signal to noise ratio (SNR) is improved by 11dB using the algorithm.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jiahuai Wang, Jisheng Han, Yong Pan, Min Zhang, Qilin Zou, Shangran Xie, "A new noise suppression algorithm for optical fiber temperature surveillance of heavy oil thermal recovery well", Proc. SPIE 8199, 2011 International Conference on Optical Instruments and Technology: Optical Sensors and Applications, 81991A (22 November 2011); doi: 10.1117/12.906604; https://doi.org/10.1117/12.906604
PROCEEDINGS
6 PAGES


SHARE
Back to Top