Shiqin He

Inaccurate prediction of shrinkage in self-compacting concrete (SCC) may result in underestimated cracking risk, increased permeability, serviceability deterioration, and reduced long-term durability of concrete structures. Although conventional empirical shrinkage models are widely used in engineering practice, their accuracy is often limited when applied to SCC mixtures with high paste volume, mineral admixtures, manufactured sand, and high-range water-reducing admixtures. Recent machine-learning-based models provide an alternative approach, but single learning algorithms may show limited robustness for small and heterogeneous datasets. In addition, random sample-level data splitting may introduce information leakage when shrinkage measurements obtained at different curing ages from the same mixture are assigned to both training and testing sets. To address these issues, this study develops a stacking-based ensemble learning framework for SCC shrinkage prediction using mixture proportions and curing age as input variables. A multi-source database containing 61 mixture designs and 448 data samples was established from published experimental studies. To obtain a more realistic assessment of model generalization, a mixture-level validation strategy was adopted, in which all age-dependent samples from the same mixture were assigned exclusively to either the training set or the testing set. Under this strategy, 358 data samples were used for model training and 90 data samples were used for independent testing. Four base learners, including multilayer perceptron (MLP), support vector regression (SVR), decision tree (DT), and gradient boosting decision tree (GBDT), were constructed and integrated through different ensemble configurations. The Stacking-SVR model achieved the best overall performance on the independent testing set, with a mean absolute error (MAE) of 13.6 με and a mean absolute percentage error (MAPE) of 7.5%. Compared with GBDT, Stacking-GBDT, and DT models, the proposed Stacking-SVR model reduced the MAPE by approximately 10.7%, 11.8%, and 35.3%, respectively. Stability and applicability analyses further indicate that the proposed framework can provide reliable shrinkage predictions within the investigated mixture and curing-age ranges. However, because the model was developed from a compiled database and does not explicitly include environmental variables such as relative humidity and temperature, its use should be limited to parameter ranges represented in the database. Overall, the results demonstrate that stacking ensemble learning combined with mixture-level validation offers a leakage-controlled and engineering-oriented approach for SCC shrinkage prediction.

Soil reinforcement using geocells and vegetation is one of the best forms of soil protection for shallow slope failure control. The geocell supports the vegetation growth and the vegetation cover provides protection against the surface erosion. This work proposed a soil treatment method using geocells for supporting the vegetation growth and stabilizing the shallow slope. A step-by-step installation of the geocells in the field and the development of vegetation growth were also described. The authors developed nine physical models that were reinforced with different sized geocell structures (no reinforcement and small and large geocell reinforcement). The models were placed under three rainfall intensities (50, 75, and 100 mm/h). The stability of the slope under the rainfall and the performance of the geocell reinforcement were assessed from the the development of slope failures, the soil erosion and the slope displacement. The results showed that the stability of geocell reinforced slopes were better off than the unsupported slope. The small geocell-reinforced slopes showed less measured soil erosion and also smaller slope displacement. In general, small geocells outperformed large geocells in terms of the erosion control and slope stabilization. The rainfall intensity dramatically increased the soil erosion on slopes. The geocell- and vegetation-treated slope in the field showed good resistance against the surface erosion.

The slump flow and flow time T200 are the main parameters to evaluate the rheological properties of cement paste. Generally, the manual measurement method is used and it is difficult to get T200 accurately. In this paper, an automatic measurement method of slump flow based on dynamic image processing technology is studied. Firstly, the cement paste flowability test video is divided into frames every 100 ms; secondly, based on the principle of perspective transformation, an automatic point selection method is proposed to correct the frame image to solve the problem of shooting angle; thirdly, the watershed algorithm based on distance transformation is used to segment the boundary of the cement paste area, and the slump flow is calculated according to the polar coordinates of the boundary; lastly, the relationship curve between slump flow and time is automatically output, and the slump flow at any time is obtained. The method is verified by nine groups of cement paste flowability tests. The results show that this method can fastly and accurately obtain the slump flow and flow time T200. Compared with the traditional manual measurement method, the measurement error of slump flow is less than 2%, and the measurement error of flow time T200 is less than 8%. The method enables real-time dynamic measurement of the slump flow.

In marine environments, the durability of reinforced concrete structures such as bridges, which suffer from the coupled effects of corrosion and fatigue damage, is significantly reduced. Fatigue loading can result in severe deterioration of the bonds between reinforcing steel bars and the surrounding concrete, particularly when reinforcing bars are corroded. Uniaxial tension testing was conducted under static loading and fatigue loading conditions to investigate the bonding characteristics between corroded reinforcing bars and concrete. An electrolyte corrosion technique was used to accelerate steel corrosion. The results show that the bond strength was reduced under fatigue loading, although the concrete did not crack. Therefore, fatigue loading has negative effects on the bond strength between corroded steel bars and concrete. The effects of corrosion cracking on bond strength become more pronounced after corrosion cracking appears along the main reinforcing bars. When the average width of cracking along main reinforcing bars exceeds 3 mm, the bonding properties deteriorate rapidly based on the effects of corrosion cracking, whereas fatigue loading exhibits no additional effects on bond strength.

There are more studies on self-compacting concrete but less on self-compacting prestressed concrete members using mechanism sand. In this paper, the stiffness and deformation performance of the test members are investigated by observing the mid-span deflection and concrete strains in different cross-sections during and after the tensioning process of 19.2-m mechanism sand self-compacting prestressed concrete beams. The test observation results are processed, and the member deflection is calculated according to the Code JTG D62-2004. Then the deflection is calculated by regressing the curvature curve of the secondary beam section through the concrete strain observations of different sections. Then the finite element model simulation is carried out to calculate the mid-span invert arch value after the tensioning process. The results are in good agreement with the actual test. Finally, the four results were compared according to the actual measured deflection values of the test. Finally, it is concluded that there are some differences between the short-term deflection calculation of the mechanism and self-compacting prestressed members using the "JTG D62-2004" and the measured results, but still can meet the design requirements the mechanism sand self-compacting prestressed members.

This work proposed a composite soil reinforcement method to support the vegetation growth and improve slope stability using geocell structures and wheat straws. Twelve physical models were developed for assessing rainfall intensities, slope steepness and geocell sizes on the soil erosion. The results showed that: (1) the geocell reinforced slopes maintained a good integrity, but the bare slope showed horizontal cracks; (2) the soil erosion was less than 40 g and more than 4000 g under the 50 mm/h and 100 mm/h rainfall intensities; the largest slope displacement was 2.5 mm and 7.5 mm, respectively; (3) the steep slopes showed a largest soil erosion of 5000 g and largest displacement of 8 cm, compared to 4500 g and 5.5 cm for gentle slopes; soil water contents were 40% and 30% for the steep and gentle slopes; (4) the measured soil erosion under the 100 mm/h rainfall was 4000 g and 5000 g for the small and large geocell reinforced slopes; the displacement of small geocell reinforced slopes were typically 1–2.5 cm smaller than large geocell reinforced slopes. It is concluded that the small geocell reinforced slope at a small inclination and low rainfall intensity has the best stability.

The durability of in‐service reinforced concrete bridges will be greatly reduced under the action of corrosion and the repeated load such as vehicles. In this paper, six reinforced concrete beams were cast and subjected to sustained load test for one year, and then, the alternating test of corrosion and fatigue load were carried out with the damaged concrete beams. The long‐term deflection, fatigue lifetime, failure modes, and crack growth were investigated under different corrosion and fatigue working conditions. The fracture section of steel bars was scanned by electron microscopy at the end of fatigue test. The results indicate that the deflection of the beams will continue to increase under a long‐term load. The chloride salt in the beam has little influence on the deflection performance, but will shorten the fatigue lifetime due to the corrosion of the steel bars. Moreover, the corrosion environment can accelerate the rusting of the beam bar and reduce the fatigue lifetime, the accumulation of deflection damage and crack damage, and other fatigue characteristics. The crack initiation‐propagation‐fracture stage of steel bars occurs in the process of fatigue.

Slope failure triggered by soil erosion under rainfall remains one of the most difficult problems in geotechnical engineering. Slope protection with planting vegetation can be used to reinforce the soil and stabilize the slope, but the early collapse of the planting soil before the complete growth of plants becomes a major issue for this method. This paper has proposed a composite soil treatment and slope protection method using the geocell structures and the wheat straw reinforcement. The geocell structures improve the stability of the planting soil and provide a stable and fixed environment for the vegetation, while the wheat straw reinforces the soil and also increases the fertility. The authors have performed a total of 9 experiments in this work that are classified into three groups, i.e., the unsupported slopes, the geocell reinforced, and the geocell and wheat straw composite reinforced with a consideration of three different rainfall intensities. The progressive slope failure development during the rainfall was assessed, as well as the soil erosion, the slope displacement, and the water content. The results show that the slope failure increases as the rainfall continues, and the soil degradation increases with the intensity of rainfall. The soil treatment using geocell improves the slope stability, but the geocell and wheat straw composite reinforcement has the best erosion control and slope protection.

In marine environments, reinforced concrete bridge structures are subjected to cyclic loads and chloride ingress, which results in corrosion of the reinforcing bars, early deterioration, durability loss, and a considerable reduction in the fatigue strength. Owing to the complexity of the problem and the difficulty of testing, there are few studies on the fatigue performance of concrete structures under the combined action of corrosion environment and cyclic load. Therefore, a coupling test device for corrosion and cyclic load is designed and fatigue tests of reinforced concrete beams in air environments and chlorine salt corrosive environments are carried out. The fatigue corrosion process, damage mode, and corrosion features of the test beams as well as chloride ion content in concrete are analyzed. The relationships of deflection, crack, and number of cycles in the different environments are given. Results show that the fatigue life of the beam is greatly reduced under coupled effects of the cyclic load and corrosive environment, the failure form of the beam is corrosion fatigue damage. The deflection and crack keeps growing with the increase in loading cycles. Under the coupling of cyclic load and corrosion environment, the content of chloride ion in concrete is low and there is less variety along the direction of penetration.

The influences of variable humidity on the creep behaviors of concrete and the long‐term deflections of RC beams are analyzed in this paper. A total of 6 prismatic specimens and 5 RC beams were subjected to sustained loads and 3 types of variable humidity conditions for 331 days. The creep strains of the prismatic specimens and the deflections of the RC beams were recorded to investigate the long‐term deformation characteristics. The test results reveal that both the creep strain and the deflection of specimens under cyclic humidity conditions exhibit approximately linear tendencies that are obviously less than those of specimens exposed only to natural air. During certain wetting cycles, the deformation rate became slower and the creep strain even recovered with an increase in the humidity, especially for shorter wetting‐drying cycles. Long‐term deflections predicted using the CEB‐FIP 90 and ACI 209R models were compared with the test results to evaluate the versatility of these current specifications. The findings included rather large errors between the predicted results and experimental data when the average relative humidity was adopted in the analytical models.