A Multi-Source Information Fusion Evaluation Method for the Tunneling Collapse Disaster Based on the Artificial Intelligence Deformation Prediction

Bo Wu,Weixing Qiu,Wei Huang,Guowang Meng,Yu Nong,Jingsong Huang

DOI: https://doi.org/10.1007/s13369-021-06359-z

IF: 2.807

2022-01-10

Arabian Journal for Science and Engineering

Abstract:The tunneling collapse is the main engineering disaster during tunnel construction by the drilling and blasting method. It has become a key issue to accurately evaluate the tunneling collapse risk. As for assessing the tunneling collapse risk and providing basic risk-controlling strategies, this research proposes a novel multi-source information fusion approach that combines the cloud model (CM), the support vector machine (SVM), and the evidence-based reasoning (ER). In the data processing phase, a variety of information sources are trained by different models to analyze the collapse risk value. The judgment from each model is then evaluated according to the model’s performance which is characterized by the reliability and importance weight. The judgments from different models are fused via evidential reasoning to give the overall collapse risk level. The novel approach has been successfully applied in the case of the Jinzhupa tunnel of the Pu-Yan Highway (Fujian, China). The results indicate that the proposed multi-source information fusion method has an evaluation accuracy of 87.5%, while the single-source information method has an accuracy of less than 70%. In addition, the predicted deformation of the surrounding rocks using artificial intelligence is utilized as a source of information to derive an advanced risk assessment. As a result, the decision-makers have a longer response time before a disaster occurs. Furthermore, the fusion model has excellent performance even if the risk result of different models has high conflict.

multidisciplinary sciences