鱼群算法优化组合核函数GPR的油井动液面预测

doi:10.12068/j.issn.1005-3026.2017.01.003

东北大学学报:自然科学版 ›› 2017, Vol. 38 ›› Issue (1): 11-15.DOI: 10.12068/j.issn.1005-3026.2017.01.003

鱼群算法优化组合核函数GPR的油井动液面预测

李翔宇¹，高宪文¹，李琨²，侯延彬¹

(1. 东北大学信息科学与工程学院，辽宁沈阳110819； 2. 渤海大学工学院，辽宁锦州121013)

收稿日期:2015-12-22 修回日期:2015-12-22 出版日期:2017-01-15 发布日期:2017-01-13
通讯作者: 李翔宇
作者简介:李翔宇(1982-)，男，辽宁沈阳人，东北大学博士研究生；高宪文(1954-)，男，辽宁盘锦人，东北大学教授，博士生导师.
基金资助:
国家自然科学基金资助项目(61573088，61433004，61403040).

Prediction for Dynamic Fluid Level of Oil Well Based on GPR with AFSA Optimized Combined Kernel Function

LI Xiang-yu¹， GAO Xian-wen¹， LI Kun²， HOU Yan-bin¹

1. School of Information Science & Engineering， Northeastern University， Shenyang 110819， China; 2. College of Engineering， Bohai University， Jinzhou 121013， China.

Received:2015-12-22 Revised:2015-12-22 Online:2017-01-15 Published:2017-01-13
Contact: GAO Xian-wen
About author:-
Supported by:
-

摘要/Abstract

摘要： 针对抽油井动液面(DFL)检测主要依靠人工操作回声仪测试，无法实时在线检测，而单一核函数的高斯过程回归(GPR)无法明显提高预测精度和泛化能力，提出了一种人工鱼群算法(AFSA)优化组合核函数的动态高斯过程回归动液面预测模型.采用多项式函数、线性函数与径向基函数组合构建核函数，利用人工鱼群算法对核函数模型参数进行寻优，采用快速傅里叶变换(FFT)和核主元分析(KPCA)融合提取时频数据非线性特征作为模型输入，提高模型的预测精度和泛化能力.油田现场应用验证了该方法的有效性.

关键词: 油井, 动液面, 人工鱼群算法, 组合核函数, 高斯过程回归

Abstract: The dynamic fluid level (DFL) of an oil well is usually measured onsite by using the acoustic method. This method， however， has its limitation in determining the real-time DFL. Considering that Gaussian process regression (GPR) with single kernel function cannot significantly improve the prediction accuracy and generalization ability， a dynamic GPR for DFL with the combined kernel function optimized by artificial fish-swarm algorithm (AFSA) was proposed. The polynomial function， liner function and radial basis function were used to construct the combined kernel function of GPR in order to improve the generalization ability. The AFSA was used to optimize parameters of the combined kernel function in order to improve the prediction accuracy. The fast Fourier transform (FFT) and kernel principal analysis (KPCA) were used to extract nonlinear features of data in the time and frequency domain as the input variables of the model. The oil field application shows the validity of the proposed method.

Key words: oil well, dynamic fluid level, AFSA (artificial fish-swarm algorithm), combined kernel function, GPR (Gaussian process regression)

中图分类号:

TP29

李翔宇，高宪文，李琨，侯延彬. 鱼群算法优化组合核函数GPR的油井动液面预测[J]. 东北大学学报:自然科学版, 2017, 38(1): 11-15.

LI Xiang-yu， GAO Xian-wen， LI Kun， HOU Yan-bin. Prediction for Dynamic Fluid Level of Oil Well Based on GPR with AFSA Optimized Combined Kernel Function[J]. Journal of Northeastern University Natural Science, 2017, 38(1): 11-15.

参考文献

[1]Li K，Gao X W，Tian Z D，et al.Using the curve moment and the PSO-SVM method to diagnose downhole conditions of a sucker rod pumping unit［J］.Petroleum Science，2013，10(1):73-80.
[2]Xing M M，Dong S M，Tong Z X，et al.Dynamic simulation and efficiency analysis of beam pumping system［J］.Journal of Central South University，2015，22(9):3367-3379.
[3]Luan G H，He S L，Yang Z，et al.A prediction model for a new deep-rod pumping system［J］.Journal of Petroleum Science and Engineering，2012，80(1):75-80.
[4]Liu Z，Wang H，Yang D.Determination of real-time dynamic fluid levels by analysis of the dynamometer card［C］// Canadian International Petroleum Conference.Calgary:Petroleum Society of Canada，2007:1-8.
[5]Yang H T，Mu L J，Zeng Y Q，et al.Real time calculation of fluid level using dynamometer card of sucker rod pump well［C］// International Petroleum Technology Conference.Kuala Lumpur:International Petroleum Technology Conference，2014:1-7.
[6]Li X Y，Gao X W，Cui Y B，et al.Dynamic liquid level modeling of sucker-rod pumping systems based on Gaussian process regression［C］// 2013 Ninth International Conference on Natural Computation (ICNC).Shenyang，2013:917-922.
[7]李翔宇，高宪文，侯延彬.基于在线动态高斯过程回归抽油井动液面软测量建模［J］.化工学报，2015，66(6):2150-2158.(Li Xiang-yu，Gao Xian-wen，Hou Yan-bin.Online dynamic Gaussian process regression for dynamic liquid level soft sensing of sucker-rod pumping well［J］.Journal of Chemical Industry and Engineering，2015，66(6):2150-2158.)
[8]王通，高宪文，蒋子健.基于黑洞算法的LSSVM的参数优化［J］.东北大学学报(自然科学版)，2014，35(2):170-174.(Wang Tong，Gao Xian-wen，Jiang Zi-jian，et al.Parameters optimizing of LSSVM based on black hole algorithm［J］.Journal of Northeastern University(Natural Science)，2014，35(2):170-174.)
[9]田中大，高宪文，石彤.用于混沌时间序列预测的组合核函数最小二乘支持向量机［J］.物理学报，2014，63(16):160508-1-160508-11.(Tian Zhong-da，Gao Xia-wen，Shi Tong.Combination kernel function least squares support vector machine for chaotic time series prediction ［J］.Acta Physica Sinica，2014，63(16):160508-1-160508-11.)
[10]魏立新，张峻林，刘青松.基于改进人工鱼群算法的神经网络优化［J］.控制工程，2014，21(1):84-87.(Wei Li-xin，Zhang Jun-lin，Liu Qing-song.Optimization of neural network based on improved fish algorithm ［J］.Control Engineering of China，2014，21(1):84-87.)

鱼群算法优化组合核函数GPR的油井动液面预测

Prediction for Dynamic Fluid Level of Oil Well Based on GPR with AFSA Optimized Combined Kernel Function

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

本文评价

[1]	张禹，何楷文，李清书，巩亚东. 面向STEP-NC自由曲面特征的加工操作方法智能决策[J]. 东北大学学报（自然科学版）, 2022, 43(7): 981-987.
[2]	蔡志辉，文光奇，韩阿康，张开华. 调质工艺对V微合金油井管用钢力学性能的影响[J]. 东北大学学报（自然科学版）, 2021, 42(10): 1414-1420.
[3]	王述红，任艺鹏，陈俊智，张紫杉. 一种改进鱼群聚类算法在结构面分组中的应用[J]. 东北大学学报:自然科学版, 2019, 40(3): 420-424.
[4]	王述红，任艺鹏，邢观华. 一种改进AFSA-Elman边坡位移预测网络[J]. 东北大学学报:自然科学版, 2019, 40(1): 115-120.
[5]	张禹，白晓兰. 基于CAFSC算法的航空发动机多管路智能布局[J]. 东北大学学报:自然科学版, 2016, 37(5): 683-687.
[6]	王通，高宪文，刘文芳. 基于改进即时学习算法的动液面软测量建模[J]. 东北大学学报:自然科学版, 2015, 36(7): 918-922.
[7]	陈晓峰;宋杰;. 量子人工鱼群算法[J]. 东北大学学报(自然科学版), 2012, 33(12): 1710-1713.