Intelligent Optimization Research on PDC Cutter Structural Parameters Based on CPO-SVR Surrogate Model

This paper proposes an intelligent optimization method for PDC cutter structural parameters based on the CPO-SVR (Crown Porcupine Optimization-Support Vector Regression) surrogate model, aiming to address the challenge of traditional reinforced cutters failing to meet the requirements of long-distance tunneling under extreme working conditions in shield machines. By rationally embedding specially shaped PDC cutters into the cutter body, the lifespan of the disc cutter and tunneling efficiency are significantly improved. The study takes five structural parameters of the PDC cutter as design variables and uses specific energy of fragmentation, average cutting force, peak impact force, and energy absorption peak as performance metrics. Cutting and impact simulations are designed, and the Plackett-Burman Design (PBD) is employed to screen key factors. Optimal Latin Hypercube Sampling is utilized to obtain sample data. Based on the CPO-SVR surrogate model, the HOA (Hippopotamus Optimization Algorithm) is integrated for intelligent iterative optimization, ultimately yielding an optimal parameter combination that balances cutting performance and impact resistance. The reliability of the results is verified through finite element simulations. This method provides an intelligent solution for PDC cutter structural design, offering significant engineering application and popularization value.