基于粒子群优化的细胞神经网络油气重力异常信息提取

doi:10.13278/j.cnki.jjuese.201503301

摘要/Abstract

摘要：

细胞神经网络方法是一种有效的重力异常提取方法,提取出的重力异常信息具有较强的横向分辨能力,但如何训练优化模板是影响该方法提取精度的关键所在。笔者引入粒子群优化算法进行参数优化,结合细胞神经网络和粒子群方法各自的特点,通过细胞神经网络动态性能分析,对模板加入约束条件,并在粒子群算法中引入收缩因子,提出一种基于改进的粒子群优化的细胞神经网络算法。使用该方法进行了模型验证和实际资料处理,处理结果表明,该方法提高了重力异常提取的稳定性,能够更准确地提取油气藏重力异常信息。

关键词: 细胞神经网络, 粒子群优化, 重力异常, 收缩因子, 油气

Abstract:

Cellular neural network is an effective method for extracting useful anomalies with a strong capability of horizontal resolution. Its key point is to optimize the template when using cellular neural network to extract hydrocarbon gravity anomaly information. We presented an improved algorithm based on the characteristics of the cellular neural network and the particle swarm optimization, joined constraint condition in template, and introduced the shrinkage factors in particle swarm optimization individual characteristics through the analysis of the dynamic performance of cellular neural network. Through the model testing and the actual data processing, the algorithm proved to be more stable to optimize the template for extracting reservoir gravity anomaly information.

Key words: cellular neural network, particle swarm optimization, gravity anomaly, shrinkage factors, petroleum gas

中图分类号:

P631.1

杜润林, 刘展. 基于粒子群优化的细胞神经网络油气重力异常信息提取[J]. 吉林大学学报(地球科学版), 2015, 45(3): 926-933.

Du Runlin, Liu Zhan. Gravity Anomaly Extraction for Hydrocarbon Based on Particle Swarm Optimization and Cellular Neural Network[J]. Journal of Jilin University(Earth Science Edition), 2015, 45(3): 926-933.

参考文献

[1] 张凤旭,孟令顺,张凤琴,等.利用Hilbert变换计算重力归一化总梯度[J].地球物理学报,2005,48(3):704-709. Zhang Fengxu, Meng Lingshun, Zhang Fengqin, et al. Calculating Normalized Full Gradient of Gravity Anomaly Using Hilbert Ttransform[J]. Chinese Journal of Geophysics, 2005,48(3):704-709.

[2] 张刚. 油气重磁异常识别及提取方法研究[D].青岛: 中国石油大学, 2011. Zhang Gang. The Method Rresearch of Identification and Extraction of Oil Gravity and Magnetic Anomalies[D]. Qingdao: China University of Petroleum, 2011.

[3] 耿喜哲,张春华. 综合小波分析与分形方法进行弱磁异常识别[J].物探与化探, 2002, 26(2): 126-130. Geng Xizhe, Zhang Chunhua. The Application of Wavelet Analysis and Fractal Technique to the Recoganion of Micro-Magnetic Anomaly[J]. Geophysical & Geochemical Exploration, 2002, 26(2): 126-130.

[4] 宋双, 刘天佑, 苏俊青. 重磁小波多尺度分解及在火山岩识别中的应用[J]. 工程地球物理学报, 2008, 5(6): 675-679. Song Shuang, Liu Tianyou, Su Junqing. The Wavelet Multi-Scale Decomposition of Gravity and Magnet and Its Application in the Identification of Volcanic Rock[J]. Chinese Journal of Engineering Geophysics, 2008, 5(6): 675-679.

[5] Chua L O, Yang L. Cellular Neural Networks:Theory[J]. IEEE Trans on Circuits, 1988,35(5): 1273-1290.

[6] Albora A M,Ucan O N,Ozmen A,et al. Septaration of Bouguer Anomaly Map Using Cellular Neural Network[J]. Journal Applied Geophysics, 2001, 46:129-142.

[7] Albora A M,Ozmen A. Wavelet Based Design of Dynamical Neural Networks for Magnetic Map Processing[R].Istanbul:University of Istanbul Press,2002:133-140.

[8] 刘展,刘茂成,魏巍,等.重力异常分离的细胞神经网络方法研究[J].中国石油大学学报:自然科学版,2010,31(1):31-34. Liu Zhan, Liu Maocheng, Wei Wei, et al. A Gravity Anomaly Separation Method Based on Cellular Neural Network[J]. Journal of China University of Petroleum: Edition of Natural Science, 2010, 31(1): 31-34.

[9] Kennddy J, Eberhart R. Particle Swarm Optimization[C]// Proceedings of IEEE International Conference on Neutral Networks. Perth: IEEE, 1995:1942-1948.

[10] Eberhart R C, Shi Y. Particle Swarm Optimization: Developments, Applicationgs and Resources[C]// Pro Congress on Evolutonary Computation 2001. Piscataway: IEEE, 2001: 81-86.

[11] 卢珊萍,于盛林.基于粒子群算法的细胞神经网络模板参数设计[J].计算机技术与发展,2009,19(4):83-86. Lu Shanping, Yu Shenglin. A Template Design Method for Cellular Neural Network Based Particle Swarm Optimizer Algorithm[J]. Computer Technology and Development, 2009,19(4):83-86.[HQ]

[12] Karakuzu C. Parameter Tuning of Fuzzy Sliding Mode Controller Using Particle Swarm Optimization[J]. International Journal of Innovative Computing Information and Control, 2010, 6(10): 4755-4770.

[13] 马国庆,黄大年,杜晓娟,等.Hartley变换在位场(重、磁)异常导数计算中的应用[J].吉林大学学报:地球科学版,2014,44(1):328-335. Ma Guoqing, Huang Danian, Du Xiaojuan, et al. Hartley Transform in the Application of the Derivatives Date[J]. Journal of Jilin University: Earth Science Edition, 2014, 44(1):328-335.

[14] 管志宁,侯俊胜.重磁异常反演的拟BP神经网路方法及其应用[J].地球物理学报,1998,41(2):242-251. Guan Zhining, Hou Junsheng. Inversion of Gravity and Magnetic Anomalies Using Pseudo-BP Neural Network Method and Its Application[J]. Chinese Journal of Geophysics, 1998,41(2):242-251.

[15] 曾琴琴,王永华,吴文贤.二维磁异常的粒子群快速成像方法及其应用[J].吉林大学学报:地球科学版,2013,43(2):616-622. Zeng Qinqin, Wang Yonghua, Wu Wenxian. Fast Imaging of 2D Magnetic Anomaly by Particle Swarm Optimization and Its Application[J]. Journal of Jilin University: Earth Science Edition, 2013, 43(2): 616-622.

[16] Chua L, Thiran P. An Analytic Method for Designing Simple Cellular Neural Networks[J]. IEEE Trans on Circuits and Systems, 1991, 38(11):1332-1341.

[17] 范正国,刘前坤,黄旭钊,等.自适应重磁对应分析[J].吉林大学学报:地球科学版,2012,42(6):1904-1919. Fan Zhengguo, Liu Qiankun, Huang Xuzhao, et al. Adaptive Correlative Analysis Between Aeromagnetic and Gravity Anomaly[J]. Journal of Jilin University: Earth Science Edition, 2012, 42(6): 1904-1919.

[18] Zarandy A, Stoeffets A K, Roska T,et al. Implementation of Binary and Gray-Scale Mathematical Morphology on the CNN Universal Machine[J]. IEEE Trans on CAS, 1998,45(2): 163-168.

[19] 侯遵泽,杨文采.中国重力异常的小波变换与多尺度分析[J].地球物理学报,1997,40(1):85-95. Hou Zunze, Yang Wencai. Wavelet Transform and Multi-Scale Analysis on Gravity Anomalies of China[J]. Chinese Journal of Sinica, 1997,40(1):85-95.

[20] 杨文采,施志群,侯遵泽,等.离散小波变换与重力异常多重分解[J].地球物理学报, 2001,44(4):534-541. Yang Wencai, Shi Zhiqun, Hou Zunze, et al. Discrete Wavelet Transform for Multiple Decomposition of Gravity Anomalies[J]. Chinese Journal of Geophysics, 2001,44(4):534-541.

[21] Hinze W J, Aiken C V,Brozena J M,et al. New Standard for Reducing Gravity Data: The North American Gravity Database[J]. Geophysics, 2005,70(4): 346-365.

[22] Wu Jianshe, Jiao Licheng. Synchronization in Dynamic Networks with Nonsymmetrical Time-Delay Coupling Based on Linear Feedback Controllers[J]. Physica, 2008, 387(3): 2111-2119.

[23] Zhong W S, Stefanovski J D,Dimirovski G M,et al. Decentralized Control and Synchronization of Time-Varying Complex Dynamical Network[J]. Physica, 2009, 142(7): 151-167.

[24] Krink T, Vesterstrom J S, Riget J. Particle Swarm Optimization with Spatial Particle Extension[C]//Proceeding of International Conference on Evolutionary Computation.[S.l.].IEEE, 2002: 1474-1497.

相关文章 15

[1]	杨烁, 丁毅, 杜发述, 祖贺川, 张君毅, 刘强, 张祺, 苏涛. 二连盆地阿南凹陷腾格尔组构造-岩性油气藏形成条件及富集模式[J]. 吉林大学学报(地球科学版), 2026, 56(3): 754-767.
[2]	王振军. 油气储层裂缝预测进展[J]. 吉林大学学报(地球科学版), 2026, 56(2): 469-482.
[3]	许中杰, 冷龙, 朱占平, 段煜, 张芳霞, 徐大志. 渤海湾盆地歧口凹陷古近纪构造背景分析：来自碎屑锆石U-Pb年代学的证据[J]. 吉林大学学报(地球科学版), 2026, 56(1): 229-251.
[4]	唐华风, 付琦, 尹永康, 高有峰, 户景松, 武海超, 田志文. 花岗岩潜山储层地质研究回顾[J]. 吉林大学学报(地球科学版), 2025, 55(6): 1806-1836.
[5]	王伟明, 胡美硕, 焦立新, 杨斌, 苗长盛, 孙宏亮, 周志超, 曲希玉, . 三塘湖盆地下石炭统新层系石油地质条件及勘探潜力 [J]. 吉林大学学报(地球科学版), 2025, 55(5): 1401-1419.
[6]	易定红, 刘应如, 李积永, 李红哲, 徐晓玲, 吴颜雄, 王爱萍, 马元琨, 涂加沙, 贾正良. 柴达木盆地南八仙地区渐新世沉积环境及储层物性控制因素[J]. 吉林大学学报(地球科学版), 2025, 55(3): 705-720.
[7]	官大勇, 张宏国, 牛成民, 王启明, 李晓辉. 渤海湾盆地辽中凹陷地区东营组湖底扇关键成藏认识与勘探实践 [J]. 吉林大学学报(地球科学版), 2025, 55(2): 351-362.
[8]	单玄龙, 徐长贵, 衣健, 牛成民, 郝国丽, 郭剑南, 闫博. 中国近海典型含油气盆地中生代岩浆活动与岩浆岩潜山油气藏[J]. 吉林大学学报(地球科学版), 2024, 54(6): 1773-1787.
[9]	徐春强, 王晨杰, 王蔚. 渤中凹陷北部中生界火山岩储层发育特征及油气成藏条件[J]. 吉林大学学报(地球科学版), 2024, 54(6): 1801-1814.
[10]	黄志, 叶涛, 肖述光, 李飞, 高文博, 戴建芳. 渤海湾盆地沙垒田西北斜坡带碳酸盐岩潜山储层形成与油气成藏[J]. 吉林大学学报(地球科学版), 2024, 54(6): 1815-1828.
[11]	刘翔宇, 单玄龙, 衣健, 苏思远, 李嘉慧, 庞赫, 邹玉洁. 渤海湾盆地渤中凹陷义县组火山喷发旋回及其分布规律[J]. 吉林大学学报(地球科学版), 2024, 54(6): 1829-1845.
[12]	何元琴, 朱红涛, 丁琳, 陶文芳, 王维, 曾智伟. 珠一坳陷古近系火山物质分类、鉴定识别方法及其深层油气地质意义[J]. 吉林大学学报(地球科学版), 2024, 54(6): 1860-1882.
[13]	郑轩, 宗兆云, 付亚群, 骆坤. 海域油气盖层叠前地震反演预测方法及应用[J]. 吉林大学学报(地球科学版), 2024, 54(6): 2128-2141.
[14]	斯尚华, 喻意, 王小龙, 王新星, 赵玉涛, 吴伟涛, . 松辽盆地三肇凹陷扶余油层致密油气充注期次及其成藏贡献 [J]. 吉林大学学报(地球科学版), 2024, 54(5): 1457-1467.
[15]	蔡潇, 夏威, 马晓东, 臧素华, 丁安徐, 刘玉霞, 花彩霞, 朱一川, 李辉. 苏北盆地溱潼凹陷古近系阜宁组页岩油储集空间特征及其影响因素[J]. 吉林大学学报(地球科学版), 2024, 54(5): 1482-1493.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed