首页> 外文会议>Geotechnical Engineering in the Information Technology Age >Parameter Estimation using a Counterpropagation Artificial Neural Network with Multiple Types of Data
【24h】

Parameter Estimation using a Counterpropagation Artificial Neural Network with Multiple Types of Data

机译:使用具有多种数据类型的反向传播人工神经网络进行参数估计

获取原文
获取原文并翻译 | 示例

摘要

We present a new subsurface characterization methodology that combines primary data and correlated secondary data with a counterpropagation artificial neural network to create parameter estimates. This methodology is applied to a subset of air permeability, compressional-wave velocity and electrical resistivity measurements that were sampled on a 2-dimensional block of Berea sandstone. Similar to traditional geostatistics, the ANN can incorporate anisotropy and uncertainty to account for patterns generally observed in the geologic setting. The ANN estimates are data-driven, can incorporate large amounts of multiple data types, do not require the computation of covariance matrices and can produce realizations in real time.
机译:我们提出了一种新的地下表征方法,该方法结合了主要数据和相关的次要数据与反向传播人工神经网络来创建参数估计值。该方法适用于在Berea砂岩的二维块上采样的空气渗透率,压缩波速度和电阻率测量的子集。与传统的地统计学类似,人工神经网络可以将各向异性和不确定性纳入考虑范围内的地质模式。 ANN估计是数据驱动的,可以合并大量的多种数据类型,不需要计算协方差矩阵,并且可以实时产生实现。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号