A novel rate of penetration optimization method considering drill string vibration for deep geological drilling process

In the process of deep geological drilling, rate of penetration is a key indicator to measure drilling cost and characterize drilling efficiency. High rate of penetration is the key to reducing drilling cost and improving drilling efficiency. However, the optimization of rate of penetration faces the nonlinear, high-dimensional changes and drill string vibration. In order to solve these problems, a novel rate of penetration optimization method considering drill string vibration is developed. First, the fuzzy C-means clustering method is introduced, which is used to identify drilling conditions and obtain the constraints for reducing drill string vibration. Then the support vector regression method is introduced to construct the rate of penetration basic model and solve the nonlinear problem. Then, a novel hybrid particle swarm algorithm (NHPSO) is designed, which improves the global search ability of the algorithm by introducing multi-weight, multi-disturbance, local search and population update mechanism. The test results based on the benchmark function of IEEE CEC2015 show that the NHPSO algorithm has good global search ability. The comparison results based on actual drilling data show that the proposed rate of penetration modeling method has higher accuracy and practical engineering application prospects.