Automatic Guidance System for Long-Distance Curved Pipe-Jacking

Pan, Guorong; Fan, Wei

doi:10.1007/s12205-020-1489-9

Cited by 6 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One is that the objective function with the average KL divergence constraint makes the smaller update step size, which results in slower learning efficiency, and the other is that the optimization problem with constraints requires higher calculated cost, and the solving process is cumbersome. For these reasons, the PPO method [27] is proposed to replace the constraint with the penalty, which can solve the problem of determining the universal penalty factor and obtain a more concise optimization form, as shown in Formula (7). Applying r t (θ) as a distance measure between the updated policy and the original policy, the application of clipping eliminates the driving force of r t (θ) exceeding [1 − , 1 + ], which limit the policy update within a certain range.…”

Section: Ppo Algorithm (1) Policy-based Frameworkmentioning

confidence: 99%

“…The laser-based guidance system [5,6] installed on TBM can measure and display TBM attitude and position deviation relative to the designed tunnel axis in real time. The TBM operators set the appropriate control parameters for TBM attitude based on the current attitude and position deviation and the rough geological information from a geological survey using their experience [7,8]. Because of the complex and ever-changing geological environment, the operators cannot obtain enough current geological information according to the geological survey and personal perception.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Real-Time Optimal Attitude Control of Tunnel Boring Machine Based on Reinforcement Learning

Jia,

Huo,

Yang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Efficient control of tunnel boring machine (TBM) tunneling along the designed tunnel axis in an unknown variable geological environment is a difficult and significant task. At present, the TBM attitude during tunneling is mostly manually controlled based on the deviation between the tunneling axis and the designed tunnel axis and their experiences. The tunneling axis from manual control is often the snakelike motion around the designed tunnel axis, even exceeding the deviation limit, for which this paper analyzed three reasons, the unknown geological environment, the hysteresis of TBM position response, and the unsolved overall optimization of tunneling axis. For these reasons, this paper proposed a real-time optimal control framework of TBM attitude based on reinforcement learning, which contains the geological information predictive model, TBM attitude and position (TBMAP) predictive model, and optimal attitude control policy (OACP). This framework can predict the current geological information in real-time and provide the corresponding real-time optimal attitude control that simultaneously considers the hysteresis of TBM position response and the overall optimization of the tunneling axis. This attitude control framework can be directly deployed to TBM without increasing costs and excessive modifications to the equipment. To verify the effectiveness of this attitude control framework, the Xinjiang Yiner Water Supply Phase II Project, using the TBM method, was adopted as a case study. The results revealed that the accuracy of geological environment recognition reached 94%, and OACP can significantly reduce the accumulated deviation of the tunneling axis from the designed tunnel axis by over 80% compared with the manual control and easily provide real-time decision support for attitude control in actual engineering.

show abstract

Section: Ppo Algorithm (1) Policy-based Frameworkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Real-Time Optimal Attitude Control of Tunnel Boring Machine Based on Reinforcement Learning

Jia,

Huo,

Yang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Long-distance curved pipe-jacking has significant engineering and economic benefits in tunneling projects crossing densely built-up urban areas, heavily trafficked road sections, and oversized cross-sections [ 1 , 2 ]. However, the traditional pipe-jacking guidance method relies on the visibility conditions of the environment and does not apply to long-distance curved pipe-jacking.…”

Section: Introductionmentioning

confidence: 99%

CEEMDAN-LWT De-Noising Method for Pipe-Jacking Inertial Guidance System Based on Fiber Optic Gyroscope

Zu,

Wang,

et al. 2024

Sensors

View full text Add to dashboard Cite

An inertial guidance system based on a fiber optic gyroscope (FOG) is an effective way to guide long-distance curved pipe jacking. However, environmental disturbances such as vibration, electromagnetism, and temperature will cause the FOG signal to generate significant random noise. The random noise will overwhelm the effective signal. Therefore, it is necessary to eliminate the random noise. This study proposes a hybrid de-noising method, namely complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN)—lifting wavelet transform (LWT). Firstly, the FOG signal is extracted using a sliding window and decomposed by CEEMDAN to obtain the intrinsic modal function (IMF) with N different scales and one residual. Subsequently, the effective IMF components are selected according to the correlation coefficient between the IMF components and the FOG signal. Due to the low resolution of the CEEMDAN method for high-frequency components, the selected high-frequency IMF components are decomposed with lifting wavelet transform to increase the resolution of the signal. The detailed signals of the LWT decomposition are de-noised using the soft threshold de-noising method, and then the signal is reconstructed. Finally, pipe-jacking dynamic and environmental interference experiments were conducted to verify the effectiveness of the CEEMDAN-LWT de-noising method. The de-noising effect of the proposed method was evaluated by SNR, RMSE, and Deviation and compared with the CEEMDAN and LWT de-noising methods. The results show that the CEEMDAN-LWT de-noising method has the best de-noising effect with good adaptivity and high accuracy. The navigation results of the pipe-jacking attitude before and after de-noising were compared and analyzed in the environmental interference experiment. The results show that the absolute error of the pipe-jacking pitch, roll, and heading angles is reduced by 39.86%, 59.45%, and 14.29% after de-noising. The maximum relative error of the pitch angle is improved from −0.74% to −0.44%, the roll angle is improved from 2.07% to 0.79%, and the heading angle is improved from −0.07% to −0.06%. Therefore, the CEEMDAN-LWT method can effectively suppress the random errors of the FOG signal caused by the environment and improve the measurement accuracy of the pipe-jacking attitude.

show abstract