Hybrid bat algorithm optimization and Semi-supervised extreme learning machine for kick detection in drilling process
作者单位：College of Information Science and EngineeringChina University of Petroleum (Beijing)
会议名称：《第32届中国过程控制会议(CPCC2021)》
会议日期：2021年
学科分类：0820[工学-石油与天然气工程] 12[管理学] 082001[工学-油气井工程] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）]
关 键 词：kick detection Semi-supervised extreme learning machine Bat algorithm Drilling
摘 要：Most existing data-driven kick detection methods are supervised machine learning techniques that presume the availability of sufficient number of kick training data samples and all data labels are *** order to get a large number of available data samples,it takes a lot of manpower and material *** solve this problem,a novel kick detection methods based Semi-supervised extreme learning machine is proposed,which can train the model by a small number of labeled samples and a large number of unlabeled ***,the performance of this algorithm is sensitive to the values of the hyper *** address this problem,we propose a novel algorithm by combining bat algorithm and semi-supervised extreme learning machine to automatically select the optimal hyper parameters of semi-supervised extreme learning machine in this ***,the field data are used for comparative experiments:the results show that the proposed algorithm can achieve a higher kick diagnosis rate in the case of less labeled samples of drilling data,which verifies the advantages and effectiveness of the proposed *** study provides a new idea for improving the detection accuracy of kick.