Modeling loess collapsibility using oedometer tests for foundation and pile settlement prediction.

Area of Science:

• Geotechnical Engineering
• Soil Mechanics
• Civil Engineering

Background:

• Loess regions in Northwest China face construction challenges due to soil collapsibility.
• Existing models for predicting loess collapse have limited engineering applicability due to reliance on single indicators.

Purpose of the Study:

• To investigate loess collapsibility characteristics in Lanzhou.
• To develop a multi-parameter model for predicting loess collapse settlement.
• To improve foundation safety in loess areas.

Main Methods:

• Laboratory collapse tests under constant pressure.
• Regression modeling to assess geotechnical parameter influence.
• Standard method for depth-wise potential collapse analysis.
• Development of an exponential function model for collapsibility coefficient prediction.

Main Results:

• Established significant relationships between collapsibility coefficient and void ratio, natural moisture content, and compression modulus.
• Identified an exponential distribution trend for potential collapse variation with depth around piles.
• Developed a comprehensive prediction model integrating five geotechnical parameters.

Conclusions:

• The developed multi-parameter model accurately predicts loess collapse settlement.
• This model offers a practical tool for risk assessment and mitigation design for foundations and pile-soil interactions in loess environments.
• Enhanced foundation safety in Northwest China's loess regions is achievable.