A novel noise suppression method for white light extrinsic Fabry-Perot interferometric fiber-optic pressure sensor in heavy oil thermal recovery downhole environment

A novel noise suppression method is proposed for the pressure surveillance of heavy oil thermal recovery well using white light extrinsic Fabry-Perot interferometric (EFPI) fiber-optic sensor. The moving grey model GM(1, 1) algorithm is applied to suppress the non-stationary noise and disturbance in downhole environment. Both theoretical analysis and field test results show that there exists a threshold effect for the moving window length selection, the optimized value of window length occurs at the threshold point which best removes the high frequency disturbance as well as retaining the fast change of actual pressure signal. Field test results show that the proposed method can increase the signal to noise ratio (SNR) of pressure profile with 10.4dB, and then can effectively suppress the large deviation in the forecasting result of daily oil production by combining temperature information.