The bearing stratum of high-rise and ultra-high-rise buildings in southwest China has inevitably faced moderately weathered red mudstone. It was a waste of the potential bearing stratum calculated according to the specification, as the bearing stratum obtained from laboratory and in situ tests was much higher than the values suggested by the specification. Rock mass surface deformation detection is of great significance in the safety management of a foundation project. Some correlation between surface deformation and failure characteristics may exist that could help to understand the bearing stratum of the moderately weathered red mudstone. This research was conducted to study the progressive failure characteristics of the moderately weathered red mudstone through surface deformation. In situ load, triaxial, and binocular visual technology were employed for data acquisition. The proposed conjecture was illustrated and verified by a group of experiments from three construction sites. Five stages could be described as the progressive failure of the moderately weathered red mudstone: compaction, elasticity, elastoplasticity, plasticity, and failure. Furthermore, the surface displacement increment fluctuates with the loading time and fades into the distance. Therefore, this research could provide a robust, practical application for analyzing the progressive failure of moderately weathered red mudstone.

No abstract

In order to improve the investigation efficiency, improve the traditional engineering investigation work mode, realize the intelligent investigation, and improve the economic and social benefits of enterprises, a red soft rock image intelligent analysis and recognition system is proposed based on deep learning methods. The intelligent recognition system includes two core algorithms: soft rock image decomposition and soft rock lithology/weathering degree recognition. The research shows that the identification model of weathering degree and the lithology identification model based on the convolutional neural network (CNN) algorithm have a good identification effect, and the identification probability of moderately weathered rock reaches 95.22% and the accuracy of lithology identification is 91.34%. With the increase in training data, the recognition effect will be further improved. The intelligent identification system has been integrated into the WeChat mini programme, App, and Web system, which can be directly applied to field geological survey operations to assist geological workers in the investigation work, and realize the intelligent identification and classification of red bed soft rock.