19 NOVEMBER 2024 WorldWide Drilling Resource® Machine Learning Can Improve TBM Wear Adapted from Information by National Institutes of Health A tunnel boring machine (TBM) is a piece of equipment uniting mechanical, electronic, hydraulic, and laser technologies for large-scale, industrialized tunnel excavation, with advantages such as a relatively steady excavation speed, shorter construction periods, a minimized ecological impact, and high efficiency. However, during construction, the harsh working environment inside the tunnel and the inherent complexity of the cutting tool system at the machine’s forefront contribute to a higher incidence of tooling system failures. During construction, the hydraulic propulsion system generates thrust, which is transmitted through the cutter plate to penetrate the rock and break it. In the process of rotary rock breaking, the wear type and degree of the disk cutter are influenced by various factors, resulting in varying levels of change. Based on the characteristics of the wear pattern, it can be classified into two types, which are normal and abnormal wear. Normal wear arises from rolling friction between the cutter ring and palm surface. Over time, the cutter ring’s diameter diminishes, while the width of the cutter edge steadily increases, and when the cutter edge width surpasses the rated value, it is considered standard for disk cutter wear failure. Abnormal wear of the disk cutter is caused by many factors, resulting in different types of wear, such as biased grinding, chipping, displacement, or the dislodgement of the cutter rings, as well as bearing damage. The probability of failure due to abnormal wear is extremely low compared to normal wear. In tunnel boring projects, wear and tear in the tooling system can have significant consequences, such as decreased boring efficiency, heightened maintenance costs, and potential safety hazards. The faults caused by disk cutter wear during the boring process can diminish boring efficiency and is challenging to detect during construction. Monitoring and addressing tool wear during construction poses certain difficulties, since exclusion of failure areas caused by disk cutter wear heavily relies on manual inspections during shutdowns, requiring downtime for handling and resulting in lower efficiency. A complete three-dimensional model of the TBM hydraulic thrust and tooling system was created to help understand how the load on the propulsion hydraulic cylinder changes as the TBM tunnelling tool wears to different degrees during construction. This enabled an in-depth exploration of the correlation between these acquired signals and the extent of the tooling system failure. The model effectively identified the failure regions, enabling timely tool replacement to avoid decreased boring efficiency under wear conditions. Machine learning introduces advanced capabilities in data analysis and prediction into TBM tool wear research and provides robust technical support to enhance TBM performance, reduce maintenance costs, and improve engineering efficiency. With the rapid development of information technology and the popularization and application of data, machine learning, as a powerful tool for data analysis and recognition, has achieved remarkable results in various fields. Machine learning algorithms can analyze large amounts of data intelligently, providing new insights for the operation and maintenance of TBMs, which improves the efficiency and quality of tunneling projects. CONST PALMER BIT COMPANY THE QUALITY GOES IN BEFORE THE RED GOES ON SIMPLY SAID - CALL FOR THE RED 800-421-2487 701-572-5271 SALES@PALMERBIT.COM WWW.PALMERBIT.COM
RkJQdWJsaXNoZXIy NDk4Mzk=