Optimal control for earth pressure balance of shield machine based on action-dependent heuristic dynamic programming

Area of Science:

• Civil Engineering
• Robotics and Control Systems
• Geotechnical Engineering

Background:

• Earth pressure balance (EPB) shield tunneling is vital for underground construction.
• Maintaining excavation face stability is critical to prevent accidents during EPB shield tunneling.
• Stable automatic control of earth pressure in the sealed cabin is a significant challenge.

Purpose of the Study:

• To propose an effective and optimal control method for earth pressure balance in EPB shield tunneling.
• To develop an adaptive control system capable of online learning during tunneling.
• To enhance the safety and efficiency of underground construction projects.

Main Methods:

• An optimal control method based on action-dependent heuristic dynamic programming (ADHDP) was developed.
• A cost function was defined based on Bellman's principle of optimality for sealed cabin earth pressure.
• The ADHDP controller architecture, including action and critic networks, was constructed for optimizing screw conveyor speed.

Main Results:

• The ADHDP-based controller successfully achieved earth pressure balance control in simulations.
• The control process demonstrated increased steadiness compared to conventional methods.
• The ADHDP controller exhibited good dynamic performance and robust anti-interference capabilities.

Conclusions:

• The proposed ADHDP method offers a stable and adaptive solution for earth pressure balance control in EPB shield tunneling.
• This approach can significantly improve the safety and reliability of underground construction.
• The ADHDP controller shows promise for real-time autonomous learning and control in complex tunneling environments.