摘要
应用吉林油区现有的实际岩石压缩系数资料对前人的经验公式进行验证的结果 ,总相对误差为 5 0 4.2 5 %～681.15 % ,由此发现这些经验公式对吉林油区并不适用。采用人工神经网络BP算法 ,以压力、孔隙度为输入层参数 ,以岩石压缩系数为输出层参数 ,分别预测了吉林油区两个地区油藏的岩石压缩系数 ,应用实际资料验证 ,相对误差仅为 12 .8% ,表明用此方法预测岩石压缩系数的可靠性。图 1表 5参
Due to the requirements of the mass balance calculation, elasticity energy calculation, well testing interpretation, and overcoming the difficulty in measuring the rock compressibility in the oil gas reservoir engineering in Jilin Oilfield,east China, earlier experience formula is tested and verified by using the present data of practice rock compressibility in Jilin Oilfield, east China, resulting in 504.20 % 681.15 % of the total relative error. The study has found that these experience formulas are hardly to be used and applied in the Jilin Oilfield. So, taking pressure and porosity as input layer parameters and rock compressibility as output layer parameter,the rock compressibility is calculated by using the error back propagation of artificial neural network,which is named as a BP method. The study has predicted the rock compressibility in the Block Ⅰ and Ⅱ of Jilin Oilfield,respectively, and tested and verified the reliability of this method by using practice data, resulting in 12.8 % of the relative error only.
出处
《石油勘探与开发》
SCIE
EI
CAS
CSCD
北大核心
2003年第4期105-107,共3页
Petroleum Exploration and Development
关键词
岩石压缩系数
相关公式
人工神经网络
BP算法
rock compressibility
relative formulas
artificial neural network
BP method
