摘要
针对智能优化控制过程中岩石可钻性参数估计存在非实时性和模型泛化能力差的问题,采用两层结构建立基于Bayesian多分支岩石可钻性估计模型。通过Bayesian分类器实现岩性分类以提高可钻性模型样本数据的相关性,细化可钻性估计模型;采用改进双链量子遗传算法优化的BPNN结构,根据不同的岩石类型建立相应的岩石可钻性IDCQGA_BPNN估计模型。结果表明,该方法通过算法优化网络模型增强了模型的泛化能力,加快了参数的估计速度和估计精度,能够满足智能优化控制过程中岩石可钻性参数估计的实时性需求。
A two-level model was established for predicting rock＇s drillability based on a Bayesian multi-branch model in order to improve the real-time calculating capability of the model and increase its generalization ability for intelligent optimization control.By using the Bayesian method for lithology classification,the correlations of different rock samples and their drillability can be refined,and consequently the rock drillability model can be improved.Using an optimized back-propagation neural network（BPNN） with an improved double-chain quantum genetic algorithm（IDCQGA）,the new model of IDCQGA_BPNN can be modified according to the lithology type of rocks.The results show that this method can not only enhance the generalization ability of the model,which is optimized by an intelligent algorithm,but also can accelerate its calculation speed and improve its accuracy.The simulation results indicate that the model is satisfied for the use in real-time intelligent optimization control process for predicting the rock drillability while drilling.
出处
《中国石油大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2014年第3期73-79,共7页
Journal of China University of Petroleum(Edition of Natural Science)
基金
陕西省自然科学基金项目(2012JQ8046)
陕西省教育厅专项科研计划(11JK0933)
