首页> 外国专利> Physical property prediction method and physical property prediction device

Physical property prediction method and physical property prediction device

机译:物理性质预测方法与物理性质预测装置

摘要

PROBLEM TO BE SOLVED: To provide an apparatus and a method having high interpretability and reliability and capable of high-speed and highly accurate prediction of physical properties. SOLUTION: This is a physical property prediction device 1 for predicting a physical property value of a compound, and a calculation unit 2 uses an electron density calculated from a wave function corresponding to the molecular structure of the compound as input data and is calculated from the molecular structure. Neural network calculation unit N10 learned with a dataset whose output data is potential, neural network calculation unit N20 learned by a dataset whose output data is the physical property value of a compound with a wave function as input data, and a prediction target. A wave function whose output obtained by inputting the electron density calculated using the wave function corresponding to the molecular structure of the compound into the neural network calculation unit N10 matches the potential calculated from the molecular structure of the predicted compound. It includes a control unit 10 that identifies and inputs the specified wave function to the neural network calculation unit N20. [Selection diagram] Fig. 3
机译:要解决的问题:提供一种具有高可解释性和可靠性的装置和方法,并且能够高速和高精度地预测物理性质。解决方案:这是用于预测化合物的物理性价格的物理性质预测装置1,并且计算单元2使用根据化合物的分子结构的波函数计算的电子密度作为输入数据计算分子结构。 Neural网络计算单元N10与数据集学到的,其输出数据是由数据集学到的神经网络计算单元N20,其输出数据是具有波函数作为输入数据的化合物的物理属性,以及预测目标。通过输入使用与化合物的分子结构的波函数输入的基函数计算的电子密度进入神经网络计算单元N10来获得的波函数与预测化合物的分子结构计算的电位匹配。它包括控制单元10,其识别并将指定的波函数识别到神经网络计算单元N20。 [选择图]图3

著录项

  • 公开/公告号JP2021189473A

    专利类型

  • 公开/公告日2021-12-13

    原文格式PDF

  • 申请/专利号JP20200090714

  • 发明设计人 椿 真史;

    申请日2020-05-25

  • 分类号G16C20/30;

  • 国家 JP

  • 入库时间 2022-08-24 22:46:27

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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