Multi-objective cellular particle swarm optimization and RBF for drilling parameters optimization.
Math Biosci Eng
; 16(3): 1258-1279, 2019 02 19.
Article
en En
| MEDLINE
| ID: mdl-30947419
ABSTRACT
Wellbore drilling parameters optimization is one of the most important issue in drilling engineering. Rate of penetration or mechanical specific energy was usually utilized as the optimization objective. The rate of penetration directly relates to the drilling cycle, while mechanical specific energy reflects the drilling efficiency. In this paper, except for rate of penetration and mechanical specific energy, the drilling life of bit is also summarized as a comprehensive assessment indicator in wellbore drilling parameters optimization problem. The drilling life of bit is taken into consideration for the design and manufacturing cost of bit compose a significant part of the drilling cost and the bit drilling life greatly influences the drilling efficiency. However, those objectives are usually related in a highly nonlinear relationship and in conï¬ict with each other. Thus, a multi-objective cellular particle swarm optimization (MOCPSO) is developed to solve the three-objective drilling parameters optimization problem. Moreover, the radius basis function (RBF) method is employed into the formation parameters identification for rate of penetration model. Performance of MOCPSO is investigated by taken a comparison with multi-objective PSO and non-dominated sorting genetic algorithm-II (NSGA-II). Effect of the four commonly used neighborhood function is also investigated by making contrasts with each other. It can be inferred that MOCPSO is statistically superior to both multi-objective PSO, NSGA-II at the 0.05 level of significance on the wellbore drilling parameters optimization problem. And the four commonly used neighborhood templates perform comparable with each other, and are not statistically different for the drilling parameters optimization problem.
Palabras clave
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Asunto principal:
Simulación por Computador
/
Diseño de Equipo
/
Industrias
Idioma:
En
Revista:
Math Biosci Eng
Año:
2019
Tipo del documento:
Article