Zhifeng Xu

This study investigates the influence of specimen size and shape on the compressive strength of mass concrete, aiming to develop a robust predictive model for large-scale structural applications, particularly in hydraulic and foundation engineering. Ordinary Portland Cement (OPC) R42.5 and three coarse aggregate gradations (20–40 mm, 40–80 mm, and 80–120 mm) with a maximum aggregate size of 150 mm were used, with water-cement ratios ranging from 0.40 to 0.50. Specimens of different sizes (cubes: 150–450 mm; cylinders: Φ150×300 mm to Φ450×900 mm) were tested at 28, 90, and 180 days. The study finds that compressive strength consistently decreases with increasing specimen size, with a more pronounced size effect in cylinders than in cubes. A multi-factor model, extending Bažant's Size Effect Law (SEL), was developed to incorporate curing age, water-cement ratio, and maximum aggregate size. The model's predictive accuracy (R² = 0.84 for cubes, R² = 0.86 for cylinders) significantly outperforms classical models such as Bažant’s SEL (R² = 0.33) and Abrams’ water-cement ratio formula (R² = 0.31). This model offers improved applicability for mass concrete with large aggregates and provides valuable insights for safety evaluations, code calibration, and the design of large-scale structures.

This study investigates the corrosion-induced mechanical degradation of Q235 steel under simulated acid rain conditions. An artificial acid rain solution, simulating environmental conditions in Hebei Province, China, was applied through accelerated electrochemical corrosion testing. Following exposure, monotonic tensile tests were conducted to evaluate changes in mechanical properties, including yield strength, ultimate tensile strength, elastic modulus, elongation, and Poisson's ratio. To understand the corrosion mechanisms, microstructural analyses using scanning electron microscopy (SEM), X-ray diffraction (XRD), and energy-dispersive spectroscopy (EDS) were performed. The results indicate that acid rain corrosion promotes the formation of surface pits and microdefects, which intensify with increasing corrosion ratio and significantly weaken mechanical performance. Yield strength, tensile strength, elastic modulus, and elongation were observed to decrease progressively with corrosion, while Poisson's ratio slightly increased. SEM images show a transition from ductile to brittle fracture patterns, associated with localised stress concentration from pitting. XRD analysis reveals early formation of α -FeOOH and Fe ₃ O ₄ , which are gradually replaced by β -FeOOH and γ -FeOOH in later stages. EDS results further identify oxygen and nitrate ions as dominant contributors to corrosion propagation. Based on the experimental data, this study proposes linear empirical models ( R ² > 0.80) to quantify the degradation of mechanical properties as a function of corrosion ratio. Compared to more complex nonlinear models, the proposed linear formulations offer comparable accuracy with improved simplicity, making them more suitable for engineering applications requiring rapid prediction. These findings provide practical guidance for the design and durability evaluation of steel structures exposed to acid rain environments.

This nondestructive technique involves obtaining the speeds of P-waves based on frequencies measured using the impact-echo method (IEM). Once the P-wave is generated, the sensor captures the horizontal displacement signal from the concrete surface. P-waves are used to estimate the compressive strength of concrete. Test results demonstrate that the velocity of P-waves is sensitive to changes in concrete strength at each age. This approach is capable of calculate the P-wave velocity with relative ease, thereby enabling the prediction of compressive strength. These measured values were used to verify the accuracy of the IEM. The proposed IEM was verified using concrete specimens produced with drilling-core for different strength grades. The results indicate that the proposed method can significantly enhance the preciseness of strength evaluation. Therefore, the P-wave based method is appropriate for in situ structures with features of portable, speed, convenience, and nondestructive of the proposed method.

High manganese steel (HMS) is extensively used in the field of liquefied natural gas applications; however, it is often susceptible to marine corrosion. This study investigates the corrosion behavior of HMS in the tropical marine environment of Sanya, China. The research encompasses both field exposure assessments and controlled laboratory experiments to derive comprehensive insights into the factors influencing corrosion rates and mechanisms in such an environment. This study aimed to simulate the corrosion behavior of HMS within a tropical marine environment by employing two primary methodologies: laboratory-based accelerated full immersion and interimmersion corrosion testing, in conjunction with actual environmental exposure tests conducted at sea. The analysis involved comprehensive macroscopic observations complemented by scanning electron microscopy, X-ray diffraction analysis, and metallographic examinations to facilitate a comparison of the results obtained from both the natural sea exposure tests and the laboratory accelerated corrosion tests. These findings imply that laboratory accelerated corrosion tests can serve as an effective alternative to actual sea exposure tests. Additionally, a time similarity coefficient was established to correlate with the findings between the natural sea exposure test and the laboratory accelerated corrosion test.

The bond-slip behavior between cold-formed thin-walled steel (CTS) and foamed concrete (FC) is a critical issue in the mechanical performance of FC-filled CTS composite wall structures. Thus, this study provides experimental and theoretical research on the bond-slip behavior between CTS and FC. A total of eleven specimens were tested using push-out configurations, considering the number of web holes, foamed concrete (FC) strength, anchorage length, and CTS section splice form. A constitutive model for bond-slip was proposed, and the regression formulas for accurately predicting the characteristic bond strength between CTS and foamed concrete were established. A finite element model was developed to investigate the bond-slip mechanism at the interface between CTS and FC. The bond-slip constitutive model accurately fits the experimental and finite element results. The results indicate that the ultimate bond strength of the specimens increases with the number of web holes; when the number of web holes reaches two, the ultimate bond strength is 155.4% of that of the non-perforated specimens. As the concrete strength increases from 3.43 MPa to 11.26 MPa, the ultimate bond strength of specimens with two web holes improves by 23.1%, while non-perforated specimens have a 54.7% enhancement. When the anchorage length is extended from 200 mm to 400 mm, the ultimate bond strength decreases by 29.3%. Additionally, when steel sections are joined in a double-span I form, the bond strength increases by 91.6% and 95.8% compared to the single-span form and the double-span box form, respectively.

Purpose: Recycled aggregate self-compacting concrete (RASCC) has the potential for sustainable resource utilization and has been widely applied. Predicting the compressive strength (CS) of RASCC is challenging due to its complex composite nature and nonlinear behavior.Design/methodology/approach: This study comprehensively evaluated commonly used machine learning (ML) techniques, including artificial neural networks (ANN), random trees (RT), bagging and random forests (RF) for predicting the CS of RASCC. The results indicate that RF and ANN models typically have advantages with higher R2 values, lower root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE) values.Findings: The combination of ML and Shapley additive explanation (SHAP) interpretable algorithms provides physical rationality, allowing engineers to adjust the proportion based on parameter analysis to predict and design RASCC. The sensitivity analysis of the ML model indicates that ANN’s interpretation ability is weaker than tree-based algorithms (RT, BG and RF). ML regression technology has high accuracy, good interpretability and great potential for predicting the CS of RASCC.Originality/value: ML regression technology has high accuracy, good interpretability and great potential for predicting the CS of RASCC.

The strength of cast-in-place concrete is obtained using the innovative pendulum hammering test method (PHTM) to reduce structural damage to concrete. A specially developed PHTM device is used to apply a fixed impact force to nails and determine the relationship between the penetration depth of nails and the compressive strength of the concrete. Adverse factors such as aggregate type and concrete moisture content on concrete strength were eliminated and their impact on PHTM test results was avoided. The compressive strength of the tested concrete cube samples ranges from 20 to 50 MPa. The reliability and repeatability of PHTM are superior to the Schmidt hammer (SRH) and pull-out testing methods used in on-site testing. Research has confirmed that PHTM is suitable for in situ testing of prefabricated buildings, beam column joints, and other densely reinforced areas, and unsuitable for coring. Compared with other testing methods, PHTM testing has high accuracy and minimal damage to concrete structures.