Study on the effects of temperature fluctuations in the seasonal frozen regions on the axial force and deformation of steel support structures in deep foundation pits

This study addresses the critical knowledge gap in understanding the complex mechanical behavior of steel support structures in deep foundation pits subjected to freeze-thaw cycles and temperature fluctuations in seasonal frozen regions, where existing design methods inadequately account for frost heave and thermal effects. Through comprehensive field monitoring and theoretical analysis of pile-anchor-steel support systems, this research reveals the significant temperature-induced effects on structural performance and develops innovative calculation methods for engineering applications. Systematic long-term monitoring demonstrates that steel support axial forces exhibit distinct three-stage evolution characteristics within the temperature range of −25.5 °C to 25.5 °C: an initial fluctuation stage with periodic minor responses, a rapid growth stage during frost heave intensification, and an attenuation-rebound stage during spring thawing. During the rapid growth stage, SSP1 increased from 249.62 kN to 335.56 kN, SSP2 from 275.09 kN to 549.56 kN, while SSP3 decreased from 931.1 kN to 874.69 kN, with the maximum sudden change amplitude reaching 39.47 % and other supports experiencing changes ranging from 10.62 % to 26.48 %. The spring thawing period resulted in maximum axial force attenuation amplitude of 98.73 %, indicating significant soil structure reconstruction effects. Based on five-element deformation coordination theory involving steel supports, support piles, soil behind piles, prestressed anchor cables, and waist beams, a novel stress analysis model considering frost heave and temperature effects was established using elastic resistance method. The model innovatively simplifies the complex multi-element system by representing soil behind piles, support piles, and prestressed anchor cables as three parallel springs. Comprehensive model validation against field measurements across two typical temperature variation periods shows theoretical calculations differ from measured values by only 2.84 %–18.84 % under various temperature conditions, demonstrating excellent computational accuracy and engineering applicability. The proposed simplified calculation method provides engineers with an effective tool for quantitative analysis of temperature stress in multiple steel supports, significantly improving design efficiency and safety. These research outcomes fill the theoretical gap in deep foundation pit support systems for cold regions and provide essential scientific foundation for technical specifications and safe underground space development in seasonal frozen regions.