Loess has the structural characteristics of porous, weakly cemented and under compacted, leading to its collapsible, disintegrative and dissolute features. To study the mechanical behaviors of basalt fiber-reinforced loess, consolidated undrained triaxial tests were carried out to investigate the effects of fiber length (FL), fiber content (FC) and cell pressure (σ₃) on the shear strength. Based on the test results, a constitutive model considering the effects of the σ₃, FL and FC was established using regression analysis, and the estimation method for the model parameters was proposed. The results show that the stress–strain curve of the unreinforced loess exhibited a strain-softening type, while the reinforced loess displayed a strain-hardening type. The peak strength of the reinforced loess was significantly higher than that of the unreinforced soil, and increased with increasing of FL, FC and σ₃. Compared with the peak strength when FL was 8 mm, the peak strength increased slightly when the FL was 12 and 16 mm, respectively. The anchoring effect and bridging effect between soil particles and fibers improved the cohesion and friction of reinforced soil, resulting in the increment in the shear strength. The experimental results are in good agreement with the model predictions, indicating that the established model and the parameter estimation method are suitable for describing the relationship between the stress and strain of basalt-fiber-reinforced loess. The research results can provide guidance of the design and construction of fiber-reinforced soil in loess areas.

The compressive strength of concrete depends on various factors. Since these parameters can be in a relatively wide range, it is difficult for predicting the behavior of concrete. Therefore, to solve this problem, an advanced modeling is needed. The aim of the literature is to achieve an ideal and flexible solution for predicting the behavior of concrete. Therefore, it is necessary to develop new approaches. Artificial Neural Networks (ANNs) have evolved from a theoretical method to a widely utilized technology by successful applications for a variety of issues. Actually, ANNs are a strong computing tool that provides the right solutions to problems that are difficult to use conventional methods. Inspired by the biological neural system, these networks are now widely used for solving a wide range of complicated problems in civil engineering. This study’’s target is evaluating the performance of developed African vulture optimization algorithm (DAVOA)-Elman neural networks (ENNs) by considering different input parameters in predicting the self-compacting concrete compressive strength. Hence, once 8 parameters and again to get as close as possible to the prediction conditions in the laboratory, 140 parameters entered to the improved version of Elman Neural Networks as input. According to the results, the element network has the lowest mean squares of the test error in predicting the compressive strength of 7 and 28 days in 100 repetitions. Further, in predicting both compressive strengths, the element grid with the Logsig-Purelin interlayer transfer function has the lowest test error, which determines the optimal transfer function. Moreover, the results showed that DAVOA as a reliable tool with time and cost savings have high power in predicting the desired characteristics. Also, in predicting both 7-day and 28-day compressive strength, networks built with 140 parameters have a 74.54 and 70.44% improvement in test error over 8-parameter networks, respectively, which directly affects this effect. Further parameters are considered as input to the network error rate in predicting the desired properties.

Water migration occurs in unsaturated loess subgrade due to extreme rainfall, making it prone to subgrade subsidence and other water damage disasters, which seriously impact road safety and sustainable development of the Loess Plateau. The study performed a rainfall test using a compacted loess subgrade model based on a self-developed water migration test device. The effects of extreme rainfall on the water distribution, wetting front, and infiltration rate in the subgrade were systematically explored by setting three rainfall intensities (4.6478 mm/h, 9.2951 mm/h, and 13.9427 mm/h, namely J1 stage, J2stage, and J3 stage), and a lateral water migration model was proposed. The results indicated that the range of water content change areas constantly expands as rainfall intensity and time increase. The soil infiltration rate gradually decreased, and the ratio of surface runoff to infiltration rainfall increased. The hysteresis of lateral water migration refers to the physical phenomenon in which the internal water response of the subgrade is delayed in time and space compared to changes in boundary conditions. The sensor closest to the side of the slope changed first, with the most significant fluctuations. The farther away from the slope, the slower the response and the smaller the fluctuation. The bigger the rainfall intensity, the faster the wetting front moved horizontally. The migration rate at the slope toe is the highest. The migration rate of sensor W3 increased by 66.47% and 333.70%, respectively, in the J3 stage compared to the J2 and J1 stages. The results of the model and the measured data were in good agreement, with the R² exceeding 0.90, which verifies the reliability of the model. The study findings are important for guiding the prevention and control of disasters caused by water damage to roadbeds in loess areas.