Shale gas drilling parameter optimization method based on improved sawtooth genetic algorithm

Abstract

The invention relates to the technical field of petroleum drilling and production engineering drilling, in particular to a shale gas drilling parameter optimization method based on an improved sawtooth genetic algorithm. The method comprises: collecting actual drilling problem parameters are collected and encoding the actual drilling problem parameters into bit strings; generating an initial population which periodically changes according to a sawtooth shape; defining a unit drilling cost as a target fitness function; carrying out sawtooth optimization population selection, crossover and mutation operators; selecting individuals with high fitness according to the target fitness function; screening individuals with high fitness to generate a filial generation population, taking the filial generation population as the reinitialized population as the start of the next iteration period, and repeating the cycle; terminating judgment when the number of repetitions reaches the set number of iterations. According to the embodiment of the invention, the population is re-initialized for many times, the population scale is periodically changed according to the sawteeth of the population algebra, the algorithm is simplified, meanwhile, the relatively strong global search capability is ensured, and the local optimization precision is improved.

Description

technical field [0001] The invention relates to the technical field of oil drilling and production engineering drilling, in particular to a method for optimizing shale gas drilling parameters based on an improved sawtooth genetic algorithm. Background technique [0002] One of the most important links in the design of drilling engineering is the optimization of drilling parameters, which has a great impact on the improvement of drilling economic benefits; the optimization of drilling parameters refers to the process of optimizing the drilling and production operations on site. Optimum combination of parameters to maximize the economic benefits of the entire project. [0003] According to the descriptions in the published literature, the correction of the Younger model with empirical formulas is relatively complex in mathematical derivation and calculation, resulting in low prediction efficiency, and the drilling rate model based on drillability representation and the real-ti...

AI Technical Summary

Problems solved by technology

[0003] According to the descriptions in the published literature, the empirical formula to correct the Younger model due to the relatively complex mathematical derivation and calculation results in low prediction efficiency, and the drilling rate model based on the drillability representation and the real-time optimization objective function of drilling parameters are not practical enough; multivariate Function extremum method and pattern search method are commonly used methods to solve parameter optimization problems in the industry

[0004] The problem with this commonly used method for solving parameter optimization problems is that the modeling process of the multivariate function extreme value method is relatively complicated, and the global search ability of the pattern search method is weak

Images