Multi-objective sidetracking horizontal well trajectory optimization in cluster wells based on DS algorithm

To optimize sidetracking horizontal well trajectory in cluster wells, a new multi-objective optimization model has been established. The new model synthetically considers the total trajectory length, the well profile energy and the accuracy of target hitting as the objective functions and takes the anti-collision with adjacent well as constraint. A novel heuristic algorithm based on DS is presented to solve the established optimization model, and the constraint is handled by a penalty method using an S-type penalty function. The comparisons between multi-objective optimization and single-objective optimization indicate that multi-objective optimization is more superior in sidetracking well trajectory optimization which can obtain a relatively short total trajectory length, a low well profile energy and an extremely high accuracy of target hitting. Furthermore, the optimization model considering anti-collision can ensure the minimum distance with adjacent well no less than the safety distance which can avoid collision in cluster wells. At last, the convergence and stability of the proposed algorithm have been discussed, and the results indicate that d-DS has good convergence and numerical stability, thus the d-DS algorithm is excellent for solving trajectory optimization problem.