Research on mud pulse signal detection based on adaptive stochastic resonance

- Stochastic resonance is used to detect original signal from noisy mud pulse signal.
- A method of using genetic simulated annealing algorithm to adaptively optimize the parameters of stochastic resonance.
- The proposed method can detect original signal and improve output SNR.
- Even the input SNR is out of detection boundary of existing methods, the stochastic resonance still has a good effect.
- We demonstrate the power and utility of proposed method using simulation and field experiment data.

The original data signals have small amplitude with a long distance signal transmission from the underground to the ground. And thousands of meters of drilling mud channel in the drilling process will contain a lot of noise sources. Therefore, the original mud pulse signals are submerged by the complex noise environment. Detecting the original signal after receiving the signal on the ground is still a serious issue. In connection with mud pulse signal detection under complex background in measurement while drilling (MWD), nonlinear bistable stochastic resonance theory and the production condition was analyzed. On the basis of this, a method of using genetic simulated annealing algorithm to adaptive optimize the parameters of stochastic resonance was proposed to realize the detection of mud pulse signal in the background of complex noise. Finally, through the simulation experiment and the field verification, this method realizes the signal detection even under the low signal-to-noise ratio (SNR). And the proposed method can overcome the shortcomings of existing detection method of signal detection performance in low SNR. It proves the proposed method's effectiveness and feasibility.