| BP神经网络在自喷井井底流压预测中的应用 |
| |
| 引用本文： | 龚晶晶,蔡忠贤,谭述.BP神经网络在自喷井井底流压预测中的应用[J].世界地质,2006,25(2):191-195. |
| |
| 作者姓名： | 龚晶晶 蔡忠贤 谭述 |
| |
| 作者单位： | 中国地质大学(武汉)资源学院,湖北,武汉,430074 |
| |
| 基金项目： | 2006年中国地质大学研究生学术探索与创新基金资助项目 |
| |
| 摘 要： | 利用神经网络具有的高度非线性映射能力,将其应用于井底流压的预测,以解决现有各种方法不能完全满足需要的矛盾。从网络的结构和算法、训练样本的选择和处理、网络的学习精度和泛化能力等三个方面对BP神经网络进行研究和改进,建立了井底流压的预测模型,并对某油田的实测井底流压数据进行了精度检验和精度分析。预测结果表明,相对误差最大为2.0821%,平均相对误差为1.5108%,绝对误差一般在0.5~1 MPa,高于其他各种计算方法的精度。
|
| 关 键 词： | 自喷井 井底流压 BP 神经网络 |
| 文章编号： | 1004-5589（2006）02-0191-05 |
| 收稿时间： | 2005-10-18 |
| 修稿时间： | 2005-10-182006-04-05 |
Application of BP neural network for prediction of flowing bottom-hole pressure in flowing well |
| |
| GONG Jing-jing,CAI Zhong-xian,TAN Shu.Application of BP neural network for prediction of flowing bottom-hole pressure in flowing well[J].World Geology,2006,25(2):191-195. |
| |
| Authors: | GONG Jing-jing CAI Zhong-xian TAN Shu |
| |
| Abstract: | Neural networks can be used in the prediction of the flowing bottom-hole pressure as it has the capability of expressing arbitrary nonlinear mapping in order to resolve the problem that the existing methods are not satisfied.By the improvement of the network's configuration and arithmetic,the training sample set's selection and deal,and the network's study precision and generalization capability,the authors construct the model for the prediction of the flowing bottom-hole pressure,also check out the precision using the actual measured flowing bottom-hole pressure data from an oilfield,and analyze the precision in the end.The prediction results indicate that the errors are less than other methods.The maximum relative error is 2.0821%,the average relative error is 1.5108%,and the absolute errors are among 0.5 MPa to 1 MPa. |
| |
| Keywords: | flowing well flowing bottom-hole pressure BP neural network |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
