In this paper, we introduce a three-dimensional triaxial apparatus with rigid walls. Its pressure chamber comprises four sliding rigid plates, a rigid specimen cap, and a rigid bottom plate. It has a three-dimensional servo hydraulic load control system, an intelligent control and data storage system, and a water-air suction control system. Considering a cuboid soil specimen as a true triaxial shear layer and a vertical principal stress transfer layer, the vertical principal stress is transferred from the transfer layer to the shear layer, and the orthogonal horizontal principal stress is applied by the horizontal slip rigid plates. That solves the technical problem of mutual interference observed in conventional three-dimensional rigid plate loading. The L-shaped loading plate is improved, which reduces the deflection and friction between them. Linear guides ensure that the horizontal stress is applied synchronously and the specimen is always centered during a test. True triaxial testing of Xi'an loess is reported, and the results confirm the applicability of the apparatus in soil testing.

Given the critical role of true triaxial strength assessment in underground rock and soil engineering design and construction, this study explores sandstone true triaxial strength using data-driven machine learning approaches. Fourteen distinct sandstone true triaxial test datasets were collected from the existing literature and randomly divided into training (70%) and testing (30%) sets. A Multilayer Perceptron (MLP) model was developed with uniaxial compressive strength (UCS, sigma c), intermediate principal stress (sigma 2), and minimum principal stress (sigma 3) as inputs and maximum principal stress (sigma 1) at failure as the output. The model was optimized using the Harris hawks optimization (HHO) algorithm to fine-tune hyperparameters. By adjusting the model structure and activation function characteristics, the final model was made continuously differentiable, enhancing its potential for numerical analysis applications. Four HHO-MLP models with different activation functions were trained and validated on the training set. Based on the comparison of prediction accuracy and meridian plane analysis, an HHO-MLP model with high predictive accuracy and meridional behavior consistent with theoretical trends was selected. Compared to five traditional strength criteria (Drucker-Prager, Hoek-Brown, Mogi-Coulomb, modified Lade, and modified Weibols-Cook), the optimized HHO-MLP model demonstrated superior predictive performance on both training and testing datasets. It successfully captured the complete strength variation in principal stress space, showing smooth and continuous failure envelopes on the meridian and deviatoric planes. These results underscore the model's ability to generalize across different stress conditions, highlighting its potential as a powerful tool for predicting the true triaxial strength of sandstone in geotechnical engineering applications.