Using supplementary cementitious materials in concrete production makes it eco-friendly by decreasing cement usage and the corresponding CO₂ emissions. One key measure of concrete’s durability performance is its porosity. An empirical prediction of the porosity of high-performance concrete with added cementitious elements is the goal of this work, which employs machine learning approaches. Binder, water/cement ratio, slag, aggregate content, superplasticizer (SP), fly ash, and curing conditions were considered as inputs in the database. The aim of this study is to create ML models that could evaluate concrete porosity. Gene expression programming (GEP) and multi-expression programming (MEP) were used to develop these models. Statistical tests, Taylor’s diagram, R ² values, and the difference between experimental and predicted readings were the metrics used to evaluate the models. With R ² = 0.971, mean absolute error (MAE) = 0.348%, root mean square error (RMSE) = 0.460%, and Nash–Sutcliffe efficiency (NSE) = 0.971, the MEP provided a slightly better-fitted model and improved prediction performance when contrasted with the GEP, which had R ² = 0.925, MAE = 0.591%, RMSE = 0.745%, and NSE = 0.923. Binder, water/binder ratio, curing conditions, and aggregate content had a direct (positive) relationship with the porosity of concrete, while SP, fly ash, and slag had an indirect (negative) association, according to the SHapley Additive exPlanations study.

As a potential replacement for traditional concrete, which has cracking and poor durability issues, self-healing concrete (SHC) has been the research subject. However, conducting lab trials can be expensive and time-consuming. Therefore, machine learning (ML)-based predictions can aid improved formulations of self-healing concrete. The aim of this work is to develop ML models that could analyze and forecast the rate of healing of the cracked area (CrA) of bacteria- and fiber-containing SHC. These models were constructed using gene expression programming (GEP) and multi-expression programming (MEP) tools. The discrepancy between expected and desired results, statistical tests, Taylor’s diagram, and R ² values were additional metrics used to assess the constructed models. A SHapley Additive exPlanations (SHAP) approach was used to evaluate which input attributes were highly relevant. With R ² = 0.93, MAE = 0.047, MAPE = 12.60%, and RMSE = 0.062, the GEP produced somewhat worse predictions than the MEP (R ² = 0.93, MAE = 0.033, MAPE = 9.60%, and RMSE = 0.044). Bacteria had an indirect (negative) relationship with the CrA of SHC, while fiber had a direct (positive) association, according to the SHAP study. The SHAP study might help researchers and companies figure out how much of each raw material is needed for SHCs. Therefore, MEP and GEP models can be used to generate and test SHC compositions based on bacteria and polymeric fibers.

Marble cement (MC) is a new binding material for concrete, and the strength assessment of the resulting materials is the subject of this investigation. MC was tested in combination with rice husk ash (RHA) and fly ash (FA) to uncover its full potential. Machine learning (ML) algorithms can help with the formulation of better MC-based concrete. ML models that could predict the compressive strength (CS) of MC-based concrete that contained FA and RHA were built. Gene expression programming (GEP) and multi-expression programming (MEP) were used to build these models. Additionally, models were evaluated by calculating R ² values, carrying out statistical tests, creating Taylor’s diagram, and comparing theoretical and experimental readings. When comparing the MEP and GEP models, MEP yielded a slightly better-fitted model and better prediction performance ( R ² = 0.96, mean absolute error = 0.646, root mean square error = 0.900, and Nash–Sutcliffe efficiency = 0.960). According to the sensitivity analysis, the prediction of CS was most affected by curing age and MC content, then by FA and RHA contents. Incorporating waste materials such as marble powder, RHA, and FA into building materials can help reduce environmental impacts and encourage sustainable development.

Ancient masonry pagodas hold significant scientific, historical, and cultural importance. However, due to the complexity of masonry materials, structures, and boundary conditions, establishing finite element static and dynamic models for ancient masonry pagodas is highly challenging. This study aimed to explore the dynamic characteristics and finite element numerical simulation methods of ancient masonry pagodas in Yongzhou, Hunan province. It focused on the Huilong Pagoda in Yongzhou, where in situ test experiments under natural excitation are conducted. The SSI and NExT-ERA methods were employed to determine the ancient pagoda’s natural frequencies, vibration patterns, and damping ratios, and to validate the NExT-ERA method. The macroscopic numerical model of the Huilong Pagoda was calibrated using measured results. Subsequently, the NExT-ERA identification results were compared and analyzed with the numerical simulation results of the dynamic characteristics. The results indicate that the first three orders of natural frequencies for the ancient pagoda in the east–west direction are 1.937 Hz, 6.802 Hz, and 21.361 Hz, respectively. Similarly, the first three orders of natural frequencies in the north–south direction are 1.935 Hz, 7.439 Hz, and 21.398 Hz. The results obtained from both methods revealed that the overall structural damping ratio ranges from 0.21% to 2.89%. The numerical model was analyzed using ANSYS, and the first three orders of natural frequencies obtained were highly consistent with the measured values, exhibiting a maximum relative error of 8.54%. The numerical simulation method developed in this study can effectively simulate masonry pagodas.

This study investigates the long-term service effects on Chinese fir (Cunninghamia lanceolata) components from ancient timber buildings in southern China. Anisotropic mechanical tests were performed to examine the evolution of mechanical properties from the perspectives of moisture absorption behavior, chemical composition, and microstructural characteristics. The results show that, after approximately 217 ± 12 years (Lvb specimens) and 481 ± 23 years (Xuc specimens) of service, the longitudinal compressive strength and corresponding elastic modulus of Chinese fir increased by about 11% and 15% and 33% and 71%, respectively, compared with fresh timber. The bending strength of the Lvb sample exhibited a slight reduction (approximately 6%), whereas the Xuc specimens showed the highest increase (33%). This difference is mainly attributed to long-term bending loads that caused structural damage in the Lvb beam specimens. In contrast, changes in lateral mechanical properties were negligible. Chemical composition analysis revealed an increase in extractive content and a reduction in cellulose and hemicellulose, leading to a notable rise in crystallinity. Scanning electron microscopy (SEM) observations further showed interlayer separation, wrinkling, and local collapse of the cell walls, suggesting significant cell wall densification. Overall, the evolution of mechanical properties is governed by the combined effects of increased crystallinity and microstructural densification, which together enhance the longitudinal and bending performance of aged timber with increasing service time. The findings provide a scientific basis for evaluating the performance and structural safety of aged timber components in the conservation of ancient timber buildings.