Numerical model for predicting rock bolt corrosion


Paul Lee, UBC; Ilija Miskovic, Norman B. Keevil Institute of Mining Engineering, University of British Columbia

Corrosion of embedded ground support elements is often undetectable by visual inspection, leading to an unexpected fall of ground. Due to the uncertain degree of corrosion, these safety hazards are usually corrected after the fact by comprehensive ground support rehabilitation program that covers any areas suspected of corrosion. Currently used ground support monitoring methods of visual inspection and pull tests are either ineffective in identifying embedded bolt corrosion or overly labour-intensive to be deployed at scale. A complementary approach is to utilize an existing geological knowledge as a basis for predicting accelerated corrosion due to galvanic interaction between steel and minerals. Field rock samples taken with a diamond hole saw can give an insight into the electrochemical properties necessary for modelling and prediction of the corrosion rate. In this study, a library of galvanic potentials for various common minerals is designed to be integrated with existing mine geological data and block models. This is supplemented by field measurements to provide the most accurate corrosion rate prediction so that preemptive rehabilitation targeting high-risk areas can be scheduled. Numerical corrosion simulation can complement existing ground support monitoring program to increase mine safety and reduce rehabilitation costs. In this presentation, we will review the preliminary results of galvanic potential measurements and demonstrate the capacity of the newly developed numerical model for predicting corrosion rate based on information that is readily available in a typical mining operation.
Mots Clés: Numerical model, simulation, galvanic corrosion, ground support, rock bolt, preemptive rehabilitation, corrosion rate prediction