首页> 外国专利> Apparatus and method for predicting product condition by using datamining in manufacturing process

Apparatus and method for predicting product condition by using datamining in manufacturing process

机译:通过在制造过程中使用数据挖掘来预测产品状态的设备和方法

摘要

The present invention relates to an apparatus and method for predicting product conditions by using data mining in the production process, which are capable of stochastically predicting a quality result of the product from causes affecting a product production result in the production process. The method includes: receiving process factors corresponding to causes affecting quality in the production process and quality variables on good products and defective products corresponding to results for the process factors to generate ontology indicating the relationship between the process factors and the quality; determining desired observation points on the basis of the generated ontology, and receiving observation coordinates and measurement data of the process factors at each observation point in real time through the production process; linearly separating non-linear values from the received observation coordinates and measurement data, and performing SVM learning based on the divided linear values; predicting the current process state from an SVM learning resu calculating an optimal process factor based on the current measurement data from the ontology and history data by using an HMM; and inducing a change in the process by providing a user with the calculated optimal process factor.
机译:用于在生产过程中使用数据挖掘来预测产品状态的设备和方法技术领域本发明涉及一种用于通过在生产过程中使用数据挖掘来预测产品条件的设备和方法,其能够根据在生产过程中影响产品生产结果的原因来随机地预测产品的质量结果。该方法包括:接收与生产过程中影响质量的原因相对应的过程因素,以及与该过程因素的结果相对应的优质产品和次品中的质量变量,以生成表示该过程因素与质量之间的关系的本体;根据生成的本体确定所需的观察点,并在整个生产过程中实时接收每个观察点的观察坐标和过程因子的测量数据;从接收到的观测坐标和测量数据中线性分离出非线性值,并基于划分后的线性值进行SVM学习;根据SVM学习结果预测当前过程状态;使用HMM,基于来自本体和历史数据的当前测量数据,计算最佳工艺因子;并通过向用户提供计算出的最佳过程因子来引起过程的变化。

著录项

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