基于人工蜂群优化支持向量机回归的 隧道塌方风险预测.

In order to predict the risk level of tunnel collapse and reduce the disaster accidents caused by tunnel collapse, a tunnel collapse risk prediction model based on artificial bee colony(ABC) optimization support vector machine regression ( SVR) was established. Firstly, considering engineering geology, hydrometeorology, design factors and construction factors, 13 main influencing factors were selected to establish the tunnel collapse risk index system. Then, the artificial bee colony algorithm was introduced to optimize the kernel parameter C and penalty parameter g of SVR, the defect of low stability of traditional SVR was solved, and the accuracy of the model was improved. In order to verify the performance of the model, the evaluation parameters of correlation coefficient (R²), mean square error (MSE) and root mean square error (RMSE) were compared and analyzed. Finally, taking a water supply project in northern Xinjiang as the research object, the tunnel collapse risk test samples were predicted, and the ABC-SVR, PSO-SVR, GA-SVR and SVR models were compared and analyzed respectively. The results show that the prediction results of ABC-SVR are 100%, PSO-SVR are 83.3%, GA-SVR and SVR are 66.67% . The prediction results of ABC-SVR are more consistent with the actual engineering results, which can provide a scientific decision-making basis for tunnel collapse risk assessment. [ABSTRACT FROM AUTHOR]