Prediction of disc cutter life during shield tunneling with AI via the incorporation of a genetic algorithm into a GMDH-type neural network

Abstract

Disc cutter consumption is a critical problem that influences work performance during shield tunneling processes and directly affects the cutter change decision. This study proposes a new model to estimate the disc cutter life (Hf) by integrating a group method of data handling (GMDH)-type neural network (NN) with a genetic algorithm (GA). The efficiency and effectiveness of the GMDH network structure are optimized by the GA, which enables each neuron to search for its optimum connections set from the previous layer. With the proposed model, monitoring data including the shield performance database, disc cutter consumption, geological conditions, and operational parameters can be analyzed. To verify the performance of the proposed model, a case study in China is presented and a database is adopted to illustrate the excellence of the hybrid model. The results indicate that the hybrid model predicts disc cutter life with high accuracy. The sensitivity analysis reveals that the penetration rate (PR) has a significant influence on disc cutter life. The results of this study can be beneficial in both the planning and construction stages of shield tunneling.