会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明申请
    • Bootstrap method for oil property prediction
    • 油料性质预测的Bootstrap方法
    • US20090119244A1
    • 2009-05-07
    • US12287673
    • 2008-10-10
    • Robert J. ChimentiPatricia H. KalamarasRobert J. PottorfBruce N. Perry
    • Robert J. ChimentiPatricia H. KalamarasRobert J. PottorfBruce N. Perry
    • G06N7/02
    • G06F19/704
    • The present invention is a method to determine models to predict physical or chemical properties of a petroleum fluid when such properties of such fluids cannot be measured by conventional analytical methods. The invention includes the steps of determining one or more models that predict one or more predetermined properties from a set of oils whose optical spectra and properties are known wherein a model corresponds to each predetermined property which is related to the spectrum, estimating the quality of the models, estimating the predictive quality for samples with unknown properties, augmenting the spectrum with shape parameters and/or properties that are predicted by the step above, and determining the quality and predictive quality of the models. The models are then used to predict properties of very small samples.
    • 本发明是当这种流体的这种性质不能通过常规分析方法测量时,确定用于预测石油流体的物理或化学性质的模型的方法。 本发明包括以下步骤:确定从一组油中预测一种或多种预定特性的模型,该组油的光谱和性质是已知的,其中模型对应于与光谱相关的每个预定属性,估计 模型,估计具有未知性质的样品的预测质量,通过上述步骤预测的形状参数和/或性质增加光谱,并确定模型的质量和预测质量。 然后使用这些模型来预测非常小的样品的性质。