As a key guarantee and cornerstone of building quality, the importance of deformation prediction for deep foundation pits cannot be ignored. However, the deformation data of deep foundation pits have the characteristics of nonlinearity and instability, which will increase the difficulty of deformation prediction. In response to this characteristic and the difficulty of traditional deformation prediction methods to excavate the correlation between data of different time spans, the advantages of variational mode decomposition (VMD) in processing non-stationary series and a gated cycle unit (GRU) in processing complex time series data are considered. A predictive model combining particle swarm optimization (PSO), variational mode decomposition, and a gated cyclic unit is proposed. Firstly, the VMD optimized by the PSO algorithm was used to decompose the original data and obtain the Internet Message Format (IMF). Secondly, the GRU model optimized by PSO was used to predict each IMF. Finally, the predicted value of each component was summed with equal weight to obtain the final predicted value. The case study results show that the average absolute errors of the PSO-GRU prediction model on the original sequence, EMD decomposition, and VMD decomposition data are 0.502 mm, 0.462 mm, and 0.127 mm, respectively. Compared with the prediction mean square errors of the LSTM, GRU, and PSO-LSTM prediction models, the PSO-GRU on the PTB0 data of VMD decomposition decreased by 62.76%, 75.99%, and 53.14%, respectively. The PTB04 data decreased by 70%, 85.17%, and 69.36%, respectively. In addition, compared to the PSO-LSTM model, it decreased by 8.57% in terms of the model time. When the prediction step size increased from three stages to five stages, the mean errors of the four prediction models on the original data, EMD decomposed data, and VMD decomposed data increased by 28.17%, 3.44%, and 14.24%, respectively. The data decomposed by VMD are more conducive to model prediction and can effectively improve the accuracy of model prediction. An increase in the prediction step size will reduce the accuracy of the deformation prediction. The PSO-VMD-GRU model constructed has the advantages of reliable accuracy and a wide application range, and can effectively guide the construction of foundation pit engineering.

Fully utilizing solid waste as supplementary cementitious materials (SCMs) while ensuring the mechanical properties of cement-based materials is one of the pathways for carbon reduction in the cement industry. Understanding the effects of the two solid wastes-limestone powder (LP) and rice husk ash (RHA) on the mechanical properties of cement-based materials is of great significance for their application in concrete. This study investigates the impact of LP and RHA on the strength of cement mortar at various ages and the microhardness of hardened cement paste. The results suggest that two materials have a certain synergistic effect on the mechanical properties of the cementitious materials. The addition of RHA effectively addresses the issues of slow strength development, insufficient late-stage strength of the cementitious material, and the low strength blended with a large amount of LP, while a suitable amount of LP can promote the strength increase in the cement-RHA system. Based on the comprehensive analysis of compressive strength and microhardness, the optimal solution for achieving high mechanical properties in composite cementitious materials is to use 10% each of LP and RHA, resulting in a 9.5% increase in 28 d strength compared to a pure cement system. The higher the content of LP, the greater the increase caused by 10% RHA in compressive strength of the composite system, which makes the strength growth rate of cementitious material mixed with 10% LP at 3–56 d 62.1%. When the LP content is 20% and 30%, the addition of 10% RHA increases the 28 d strength by 44.8% and 38.8%, respectively, with strength growth rates reaching 109.8% and 151.1% at 3–56 d.

To reduce the mining of high-grade magnesite and solve the environmental pollution caused by magnesite tailings, magnesite tailings were used to produce MgO expansion agent (MEA), and a detailed study of its performance was carried out in this study. Firstly, the effects of different calcination times on the calcination products, the specific surface area, and the activity of MEA were analyzed. Then, the MEA produced by calcinating at 950 °C for 1 h was taken as the research object, and the effects of its content on the expansion performance, compressive strength, and flexural strength of the mortar were studied. The results showed that the decomposition of magnesite tailings after high-temperature calcination produced MEA, and the longer the calcination time, the lower the activity. The calcined tailings could compensate for the shrinkage of the mortar, and the expansion increased with the increase in curing temperature. What is more, when the content was less than 8%, the hydration of MEA filled the pores and improved the compactness, so the strength of the mortar increased with the increase in the expansion agent content. When the dosage was greater than 8%, excessive expansion increased the porosity, causing harmful expansion of the mortar and damaging its integrity, leading to a decrease in strength. Fly ash reduced the expansion of mortar, and after adding 30% fly ash, the expansion decreased by 20.0–36.1%, and the ability to suppress expansion decreased with the increase in curing temperature.

This study comprehensively evaluates the mechanical properties, shrinkage behavior, and durability of concrete prepared with limestone- and granite-manufactured sands under standard-curing and steam-curing conditions. The results indicate that limestone-manufactured sand concrete consistently exhibits superior compressive strength and splitting tensile strength across all curing ages, outperforming granite-modified counterparts. The introduction of granite-manufactured sand significantly degrades these mechanical properties, with deterioration intensifying as granite content increases. Dynamic elastic modulus and damping ratio analyses reveal that limestone-based concrete maintains the highest dynamic stiffness and lowest energy dissipation under both curing regimes, suggesting fewer internal defects. In contrast, granite incorporation reduces the dynamic elastic modulus and increases the damping ratio, reflecting structural deterioration and enhanced energy loss. Drying shrinkage tests demonstrate that limestone concrete achieves the lowest shrinkage deformation throughout the testing period, even under steam-curing conditions. Conversely, granite addition markedly elevates shrinkage, particularly under steam-curing conditions, leading to compromised volumetric stability. Durability assessments highlight that manufactured sand concrete exhibits higher capillary absorption, electrical flux, and porosity, attributed to inherent material defects and the surface characteristics of manufactured sand. Granite-modified concrete further weakens interfacial shear strength between aggregates and cement paste, indicating poor interfacial bonding. Steam curing exacerbates microstructural defects, emphasizing the need to optimize curing protocols. The findings propose strategies for enhancing manufactured sand concrete performance: improving interfacial adhesion between aggregates and cement paste, rationalizing supplementary material dosages, and refining steam curing regimes. These measures offer potential pathways to develop high-performance manufactured sand concrete with balanced mechanical and durability properties.