Improving Rate-of-Penetration and Reducing State Fluctuation of Drill-String System Based on Multiobjective Optimization

This article is concerned with the balance between improving rate-of-penetration and suppressing drill-string vibrations by optimizing weight-on-bit and rotational speed. The main contribution of this article lies in three points. First, due to the lack of downhole data, the problem of suppressing drill-string vibrations is transformed into reducing state fluctuation of drill-string system. Meanwhile, a multiobjective optimization scheme is established. Second, the temporal relations are revealed, including modeling time, optimization time, resampling time, controller response time, buffer time, and optimization interval. They act as constraints for subsequent modeling and optimization. Third, the improved extreme learning machine methods are developed to build the calculation models of rate-of-penetration and state fluctuation of drill-string system, separately. The nondominated sorting genetic algorithm II, technique for order preference by similarity to an ideal solution method, and moving window are used to obtain the optimal weight-on-bit and rotational speed. The effectiveness of the approach is demonstrated through an industrial case study based on the actual drilling data.