基于径向基过程神经网络的油田开发指标预测  

Oilfield Development Indicators Prediction Based on Radial Basis Process Neural Network

作　　者：许少华[1] 毕聪聪 张宇[1] 王春艳[1] 

机构地区：[1]东北石油大学计算机与信息技术学院,黑龙江大庆163318

出　　处：《计算技术与自动化》2015年第3期52-54,共3页Computing Technology and Automation

摘　　要：目前为止,现有的油田开发指标预测方法难以反映实际存在的时间累积效应对该指标预测的影响。因此,为提高油田开发指标预测的准确度,本文提出基于径向基过程神经元网络的油田开发动态指标预测模型,并将其应用到实际油田开发动态指标的预测中。实例分析结果表明,本文提出的径向基过程神经元网络的油田开发动态指标的预测方法精度高、速度快,是预测油田开发指标的一种较实用的方法。Because the existing methods are difficult to reflect the effect of actual existence of time accumulation on oil-field development indicators prediction,so in order to improve the forecast accuracy,this paper presented a model of oilfield development indicators prediction based on radial basis process neural network,which was applied to the actual dynamic oil-field development indicators prediction.Example analysis shows that the proposed method has high precision and fast speed. So it is a more practical method for prediction of development indicators of oilfield.

关 键 词：油田开发 径向基过程神经元 动态指标 预测 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]