Abstract: To explore the coupling relationship between key geological parameters of rock mass and the working parameters of tunnel boring machines (TBMs), a multidimensional association algorithm is developed to establish a coupling model of TBM based on the staticmapping relationship of rock parameters. The Kmeans clustering method is used to determine the coupling relationship between geological and rock conditions, as well as the tunneling speed during the service process. After class coding, seven main parameters, including tensile strength, compressive strength, peak slope index, weak face spacing, tunneling speed, alpha angle, and rock type, are selected to analyze and summarize the rules of shield tunneling. Compared with the decision tree, the rules are more abundant and intuitive, making them suitable for early decisions. As an effective datamining method, the association algorithm can provide a theoretical reference for intelligent tunneling.

Key words: tunneling, rockmachine information perception, data mining, association algorithm, intelligent decision