EUVIMEDEUVIMED entdecken
Crossref

Improved Variational Bayesian Adaptive Kalman Smoother Method for Random Noise Suppression in Continuous-Wave Mud-Pulse Telemetry

Xueyang Zhao, Zhidan Yan, Ruirui Sun, Fangyun Du, Dongxiao Pang, Zuodan Wang, Zixiang Chen

SPE Journal · 01.11.2026 · 1-23

DOI: 10.2118/237003-pa ↗

Kurz gefasst

Fachartikel in der Zeitschrift SPE Journal, online erschienen am 09.11.2026. Verfasst von Xueyang Zhao und 6 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Continuous-wave mud-pulse telemetry has been widely used in measurement-while-drilling (MWD) to meet the increasing demand for large-volume well-to-surface data transmission because it offers high stability and a transmission rate exceeding 4 bits/sec. However, during uplink transmission through the wellbore channel, mud-pulse signals are susceptible to pump noise and broadband random noise, which seriously compromises reliable signal identification and decoding. To address this issue, we propose an improved variational Bayesian adaptive Kalman smoother (IVBAKS) method based on a continuous-wave signal model. The method integrates variational Bayesian (VB) inference for adaptive measurement-noise estimation and Rauch-Tung-Striebel (RTS) Kalman smoothing for improved state estimation. Its main improvement is the introduction of automatic relevance determination (ARD), which converts model-parameter uncertainty into regularization factors in the filtering and smoothing stages, thereby reducing error propagation caused by model mismatch and parameter errors during backward smoothing. Simulation signals with random-noise signal-to-noise ratios (SNRs) ranging from −5 dB to −25 dB are constructed, and the proposed method is compared with the variational Bayesian adaptive Kalman filter (VBAKF), variational Bayesian adaptive Kalman smoother (VBAKS), and the non-Kalman stochastic-resonance-based estimation-of-distribution algorithm (EDA)-cascade second-order tri-stable stochastic resonance (CSTSR) method. In addition to single-trial comparisons, 30 Monte Carlo trials, carrier-frequency drift and amplitude-variation tests, an ARD ablation test, and a process noise covariance Q sensitivity analysis are performed. The results demonstrate that IVBAKS provides a good overall balance among waveform recovery, bit error rate (BER), and computational cost. Under the representative −15 dB condition, IVBAKS achieves the highest average correlation coefficient (0.774) and an average BER of 0.08 in the Monte Carlo tests with an output SNR of approximately 3.60 dB. The method also remains effective under the tested carrier-parameter mismatch range. Ground hydraulic circulation data further verify that the combined scheme of standard Kalman pump noise suppression and IVBAKS random noise suppression can effectively recover binary frequency shift keying (BFSK) pulse waveforms. Surface-acquired data from three different wells at approximately 1,000 m, 1,500 m, and 2,000 m further show that IVBAKS enhances the recoverability of BFSK carrier signal components and achieves a BER of less than or equal to 0.02 for the evaluated standard sequence segments. The results indicate that IVBAKS can improve weak continuous-wave mud-pulse signal recoverability under the tested noise and field conditions.

Zitieren

APA

Zhao, X., Yan, Z., Sun, R., Du, F., Pang, D., Wang, Z., & Chen, Z. (2026). Improved Variational Bayesian Adaptive Kalman Smoother Method for Random Noise Suppression in Continuous-Wave Mud-Pulse Telemetry. SPE Journal, 1-23. https://doi.org/10.2118/237003-pa

Vancouver

Zhao X, Yan Z, Sun R, Du F, Pang D, Wang Z, et al. Improved Variational Bayesian Adaptive Kalman Smoother Method for Random Noise Suppression in Continuous-Wave Mud-Pulse Telemetry. SPE Journal. 2026:1-23. doi:10.2118/237003-pa

BibTeX
@article{Zhao2026,
  author = {Zhao, Xueyang and Yan, Zhidan and Sun, Ruirui and Du, Fangyun and Pang, Dongxiao and Wang, Zuodan and Chen, Zixiang},
  title = {Improved Variational Bayesian Adaptive Kalman Smoother Method for Random Noise Suppression in Continuous-Wave Mud-Pulse Telemetry},
  journal = {SPE Journal},
  year = {2026},
  pages = {1--23},
  doi = {10.2118/237003-pa}
}
Kostenlos starten →