基于遗传算法优化BP神经网络在原油产量预测中的应用:以大庆油田BED试验区为例  

BP Neural Network Based on Genetic Algorithm Applied in Crude Oil Production Forecast:Taking the BED test area of the Daqing oilfield as an example

作　　者：马林茂[1] 李德富[1] 郭海湘[1] 李伟伟[1] 

机构地区：[1]中国地质大学(武汉)经济管理学院,湖北武汉430074

出　　处：《数学的实践与认识》2015年第24期117-128,共12页Mathematics in Practice and Theory

基　　金：国家自然科学基金(71103163);教育部人文社会科学研究青年基金(10YJC790071);中央高校基本科研业务费专项资金(CUG120111;CUG110411;G2012002A);构造与油气资源教育部重点实验室开放课题(TPR-2011-11)

摘　　要：油田产量预测工作一直是油田开发中的一项重要工作,许多传统的回归模型以及智能算法都已经在油田产量预测中有了应用.虽然神经网络以其较强的非线性拟合能力.而得到广泛应用,但是传统BP神经网络容易陷入局部最优值而影响预测结果.将利用遗传算法同时优化BP神经网络连接权值和阈值的算法应用到大庆油田BED试验区高含水阶段的油田产量预测,结果表明在面对高含水阶段更加复杂的地质条件和数据波动更强的情况下优化后的神经网络收敛速度更快而且预测精度更高.Oilfield production forecast has been an important work in the oil field devel- opment, many traditional regression model and intelligent algorithms have been widely used for oilfield production forecast. Neural network with its strong nonlinear fitting has been extensively adopted, but the traditional BP neural network is easy to fall into local opti- mum thus lower the accuracy of the prediction. BP neural network combined with Genetic algorithm is proposed, The BP neural network which connection weights and threshold has been optimized by Genetic algorithm is applied to the production forecast of the test area of the oilfield of Daqing Oilfield in high water cut stage, the results show that the neural network optimized by genetic algorithm converges to steady state faster and demonstrates higher prediction accuracy than the traditional neural network in the case of more complex geological conditions and more data fluctuation in the high water cut stage.

关 键 词：遗传算法 BP神经网络 油田产量预测 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] O212.1[自动化与计算机技术—控制科学与工程]