摘要
压裂实时监测是水力压裂效果评价和工程参数优化的重要保障措施之一。传统电磁监测裂缝参数识别方法的准确性和精度难以保证,影响了监测效果。为了提高裂缝参数识别能力,提出了一种基于海洋捕食者密度聚类混合优化算法的电磁监测裂缝参数识别方法。利用海洋捕食者算法(MPA)进行多次寻优,以每次寻优结果作为初始数据集,然后,利用密度聚类算法(DBSCAN)进行聚类,构建中间样本数据集,最后,抽取该样本数据中值作为最终输出结果。采用Rastrigin函数进行测试,分析混合优化算法寻优能力。测试结果表明,相对粒子群优化算法(PSO),MPA算法单次寻优效果较佳。但两种算法寻优结果均具有较强随机性,其中,PSO和MPA算法50次寻优精度分别为10^(-7)~10^(2)和10^(-10)~10^(-2),而改进的混合优化算法寻优效果更稳定,寻优精度达10^(-7)。构建缝长、方位压裂模型并进行了数值模拟实验,结果表明,在噪声低于15%时,缝长和方位识别平均绝对误差分别小于1 m和1°。利用改进的算法对四川盆地某井页岩气压裂电磁监测实测数据进行分析,确定了裂缝改造的长度(缝长)与方位。实例分析结果验证了改进算法的可行性和有效性。
Real-time monitoring of fracturing is an important supporting measure for hydraulic fracturing effect evaluation and engineering parameter optimization.In traditional electromagnetic monitoring,the accuracy and precision of fracturing parameter identification can hardly be ensured,which greatly affects the monitoring results.In this paper,a method of fracturing parameter identification in electromagnetic monitoring based on marine predator-density clustering hybrid optimization algorithm was presented to improve the ability of fracturing parameter identification.Specifically,the marine predator algorithm(MPA)is employed for multiple optimizations,and the outcomes of each optimization are extracted as the initial dataset.Then,the density-based spatial clustering of applications with noise(DBSCAN)is used for clustering,and the intermediate sample dataset is constructed.Finally,the median value of the sample data is extracted as the final output result.The optimization ability of the hybrid optimization algorithm is analyzed through the Rastrigin function test.The test results show that compared with the particle swarm optimization(PSO)algorithm,MPA has a better single optimization effect,but the optimization results of both algorithms have strong randomness,with the 50-optimization accuracy of 10^(-7)~10^(2) and 10^(-10)~10^(-2),respectively.In contrast,the hybrid optimization algorithm exhibits more stable result,with the optimization accuracy up to 10^(-7).The fracture model of fracture length and azimuth is built and numerical simulation experiment is carried out.The results show that the average absolute error of fracture length and azimuth identification is less than 1 m and 1°respectively under the influence of less than 15%noise.Taking the measured data of electromagnetic monitoring of shale gas fracturing in a well in the Sichuan Basin as an example,the length and azimuth of hydraulic fracture are determined.The research results have certain guiding significance for electromagnetic monitoring of hydraulic
作者
曾波
杨扬
宋毅
陈珂
徐尔斯
王怡亭
徐颖洁
裴婧
ZENG Bo;YANG Yang;SONG Yi;CHEN Ke;XU Ersi;WANG Yiting;XU Yingjie;PEI Jing(Shale Gas Research Institute,PetroChina Southwest Oil&Gasfield Company,Chengdu 610051,China;Sichuan Changning Natural Gas Development Co.,Ltd.,Chengdu 610066;School of Geosciences and Info-Physics,Central South University,Changsha 410083,China)
出处
《石油物探》
CSCD
北大核心
2024年第3期684-693,共10页
Geophysical Prospecting For Petroleum
基金
中国石油西南油气田页岩气研究院合作项目(XNS页岩院JS2021-39)资助。
关键词
水力压裂
电磁法
海洋捕食者算法
密度聚类算法
实时监测
裂缝参数识别
hydraulic fracturing
electromagnetic method
marine predator algorithm(MPA)
density-based spatial clustering of applications with noise(DBSCAN)
real-time monitoring
fracturing parameter identification
