Freeze-thaw damage is a significant factor affecting the durability of concrete in cold regions. The research on the mechanisms of freeze-thaw damage in concrete remains inconclusive. Currently, the primary methods for investigating freeze-thaw issues in concrete include experimental research and numerical simulation studies. The research focus can be categorized into two areas: macro structural performance degradation and micro pore structure analysis. However, most current value methods are limited to the micro level when simulating the freeze-thaw behavior of concrete, and there is a lack of research on the macroscopic mechanical properties of concrete following freeze-thaw cycles. At the same time, grid dependence arises when addressing crack problems. Consequently, this work proposes a mesoscopic-scale peridynamic freeze-thaw model. The relationship between pore frost heaving force and frost heaving force state is established. The pore frost heaving force is incorporated into the peridynamic motion equation as a force state, enabling the simulation of cracks and the prediction of residual bearing capacity in freeze-thaw damaged concrete.

The commercial concrete purchased by Qingdao Yehua Jianzhong Construction Engineering Co., Ltd. is used. The concrete size is 100 mm × 100 mm × 100 mm, the concrete strength is C40, and the water-cement ratio (W/Cs) is 0.51. The freeze-thaw cycle test was carried out by TDR-15D concrete rapid freeze-thaw test machine. The number of freeze-thaw cycles was set to 0, 30, 60, 90 times and 120 times, and the temperature range was (-18±2)-(18±2) ℃. Three concretes were set up for repeated tests under each cycle. Before the start of the test, the concrete was first immersed in water for 4 d to ensure that the water fully penetrated the concrete. Then the freeze-thaw test machine was started, and the freeze-thaw cycle test was carried out according to the preset cycle times and temperature. After the test, the concrete test block was taken out to dry the surface scum and water, numbered and weighed, and the dynamic elastic modulus was detected. The uniaxial compression test was carried out on the MTS Exceed E64.305 electro-hydraulic servo universal testing machine.

The open porosity measured by the test is 1.68%, and the pore frost heaving force calculated by the finite element method is 58.3 MPa. After obtaining the relevant parameters, the freeze-thaw simulation is realized by matlab programming based on the peridynamics theory.

The results of the freeze-thaw tests and uniaxial compression tests were compared with the simulation outcomes. The diagram illustrating simulated freeze-thaw damage depicts the phenomenon of surface spalling and internal crack propagation, which aligns with the test results. The uniaxial compression stress-strain curves exhibit a similar trend, with a maximum difference in peak stress of 4.7%. After undergoing freeze-thaw cycles, the elastic modulus of the test group was initially low but subsequently increased. This is due to the cracks being filled with ice and broken concrete chips, resulting in a loose state of the concrete, which leads to a low elastic modulus during the initial stages of loading. As the load increases, the stiffness of the concrete also increases once the crack has closed.

The open porosity and pore frost heaving forces of concrete are critical factors influencing its freeze-thaw performance. The freeze-thaw damage and residual bearing capacity of concrete with open porosities of 1.0%, 3.0%, 5.0%, 7.0% and 9.0%, as well as pore frost heaving forces of 40, 50, 60, 70 MPa and 80 MPa, are discussed in detail. The influence of open porosity and pore frost heaving forces on the freeze-thaw performance of concrete follows a similar pattern. When the two values are low, there are no freeze-thaw cracks or minor freeze-thaw cracks in the concrete. When the two values are large, the freeze-thaw cracks cover the whole concrete. The rate of decrease in the residual bearing capacity of concrete initially increases and then subsequently decreases. This is because when the freeze-thaw damage is small, each new damage may become the starting point of the freeze-thaw crack, so the damage will promote the new damage. The damage caused by freezing and thawing in the later stages reaches saturation, resulting in a limited number of cracks and overall damage in the concrete. The concrete has been damaged and should be regarded as having completely lost its load-bearing capacity.

The main conclusions of this paper are summarized as following. The peridynamic freeze-thaw model closely resembles the test used to simulate surface spalling and crack propagation in concrete subjected to freeze-thaw cycles. The simulation results of uniaxial compression for freeze-thaw concrete align closely with the experimental findings. The maximum difference in stress values is 4.7%, while the maximum difference in residual bearing capacity is 3.7%. The rate of freeze-thaw damage in concrete increases gradually during the early stages and then slows down in the later stages. The freeze-thaw damage of concrete is characterized by surface damage spalling and internal crack staggered expansion. Through regression analysis, the linear relationship between freeze-thaw damage and residual bearing capacity is established, with a coefficient of determination reaching 0.966. Compared to the mechanical model calculations, the regression curve is utilized to predict the residual bearing capacity based on freeze-thaw damage, resulting in an efficiency increase of approximately 360 times.