Stability classification probability model of loess deposits based on MCS-Cloud.

The stability classification of loess deposits around tunnels is a vital prerequisite for safe construction in underground environment. Due to the fuzziness and randomness of loess physical and mechanical parameters, the stability prediction of loess deposits shows uncertainty. Existing loess deposit stability classification models rarely consider the uncertainty of influencing factors. A novel classification probability model of loess deposits is proposed for the above problems based on Monte Carlo simulation and multi-dimensional normal cloud (MCS-Cloud). Specifically, five loess parameters, including water content, cohesion, internal friction angle, elastic modulus, and Poisson ratio, were selected as predictors for the stability level of loess deposits. The weights of the predictors were obtained through 50 test samples. After acquiring the numerical characteristics of the normal cloud, the stability level can be comprehensively evaluated with the weighted multi-dimensional normal cloud model. The classification model was applied to the loess tunnel in Yan'an, China. The prediction results are in good agreement with practical engineering, denoting the rationality of the weighted multi-dimensional normal cloud. Finally, the stability classification of loess deposits was discussed from the perspective of uncertainty analysis with the application of MCS. Results proved that the MCS-Cloud model is feasible for classifying the stability of loess deposits surrounding tunnels. The obtained classification probability can be used for quantitative risk assessment of loess tunnels.

Intelligent prediction of rockburst in tunnels based on back propagation neural network integrated beetle antennae search algorithm.

Rockburst is one of the major engineering geological disasters of underground engineering. Accurate rockburst intensity level prediction is vital for disaster control during underground tunnel construction. In this work, a hybrid model integrating the back propagation neural network (BPNN) with beetle antennae search algorithm (BAS) has been developed for rockburst prediction. Before model building, 173 groups of rockburst dataset were collected. Six geological parameters are selected as predictors for rockburst, including the maximum tangential stress of the surrounding rock σθ, the uniaxial compressive strength of rock σc, the tensile strength of rock σt, the stress ratio σθ/σc, the rock brittleness ratio σc/σt, and the elastic energy index Wet. After preprocessed by outlier detection and synthetic minority oversampling technique (SMOTE), the new dataset was divided into training and test parts. BAS could optimize the weights and biases of BPNN from the training process. Then the established hybrid model was applied to the test samples with predicted accuracy of 94.3%, proving that the hybrid model has practical value in researching rockburst prediction.