Yong Tan

Precise simulation of rainfall is crucial for physical modeling of rainfall-related geohazards and geotechnical failures. The practicality of conventional rainfall simulators was often compromised by several limitations, such as restricted real-time adjustability, significant cost, and a lack of versatility across different testing needs. This study introduced the development and implementation of a cost-effective, modular rainfall control system designed to simulate complex rainfall patterns in geotechnical model testing. The system integrated Arduino Uno boards to establish a dual-mode control framework, enabling manual adjustment for controlling parameter calibration and automated regulation for programmable rainfall scenarios. Employing pulse-width modulation (PWM) for pump control and a Hall-effect sensor for flow monitoring, the system provided finer, real-time modulation of rainfall intensity. It reliably achieved rainfall intensities from 74 to 440 mm/h, with a required flow stabilization time of 8 to 15 s after each adjustment. Under the system’s control, the custom-designed nozzle array delivered highly uniform rainfall, maintaining a Christiansen’s uniformity coefficient above 90 %. The system was subsequently validated through a series of reduced-scale 1 g physical model tests that simulated rainfall-related seepage erosion in a braced excavation. Test results confirmed that the system not only controlled rainfall intensity variations effectively in accordance with predefined temporal patterns but also achieved spatially uniform rainfall distribution across the model surface. Due to its scalability and cost-effectiveness, the control system supports a wide range of geotechnical model setups requiring precise and repeatable simulation of time-dependent rainfall conditions.

This study presented a general analytical solution for calculation of lateral active earth pressure of narrow backfills behind retaining walls under arbitrary lateral displacement modes. The proposed method innovatively introduced an equivalent friction angle between adjacent horizontal differential elements to characterize the influence of diverse wall displacement modes on stress redistribution within the backfill soils, addressing three critical gaps: (1) universal applicability—a single framework replaced mode‐specific formulations, enabling predictions for any wall displacement mode, including complex drum‐shaped deformations in braced excavations; (2) maximum displacement localization sensitivity—a nonlinear transition law for explicitly accounted for the depth‐dependent influence of maximum wall deflection (), enhancing accuracy for flexible walls; and (3) simplified workflow—the solution required only basic geotechnical and geometric parameters (, , ) and avoided iterative steps, ensuring efficiency for practical design. Validation against finite element simulations and published model tests demonstrated strong agreement across displacement modes. Parametric analyses revealed that the proposed method effectively captured the sensitivity of earth pressure to backfill width (), internal friction angle (), and wall‐soil friction angle (). Notably, when , the solution predicted significantly reduced earth pressures compared to classical theories, emphasizing its advantages in optimizing retaining structure designs for urban excavations adjacent to existing underground facilities.