Data on evolutionary hybrid neural network approach to predict shield tunneling-induced ground settlements

Area of Science:

• Geotechnical Engineering
• Civil Engineering
• Earth Sciences

Background:

• Shield tunneling operations can induce significant ground settlements.
• Accurate prediction of these settlements is crucial for urban infrastructure safety.
• Karst geological conditions present unique challenges in tunneling projects.

Purpose of the Study:

• To present a comprehensive dataset of shield tunneling-induced ground settlements.
• To include geological and shield operation parameters influencing ground settlement.
• To incorporate data on karst features and their treatment schemes.

Main Methods:

• Field monitoring of ground settlements during shield tunneling.
• Tabulation of 17 key variables: geological conditions and shield operation parameters (time series).
• Inclusion of specific karst parameters: height, distance to tunnel invert, and treatment methods.

Main Results:

• A detailed dataset of ground settlements and influencing factors is provided.
• The dataset captures the dynamic nature of shield operation parameters.
• Specific data on karst interactions with tunneling is documented.

Conclusions:

• The dataset enriches existing knowledge on shield tunneling impacts.
• It serves as a valuable resource for developing AI-driven settlement prediction models.
• This data is foundational for research on tunneling in complex geological settings, including karst terrains.