...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of Intelligent Manufacturing >Online incremental learning for tool condition classification using modified Fuzzy ARTMAP network
【24h】

Online incremental learning for tool condition classification using modified Fuzzy ARTMAP network

机译：使用改进的Fuzzy ARTMAP网络的工具条件分类在线增量学习

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Condition monitoring of tool wear is paramount for guaranteeing the quality of workpiece and improving the lifetime of the cutter. To improve the training speed and the flexibility of the incremental learning, a modified Fuzzy ARTMAP classifier is developed in which the resonance layer is linked with the category node directly by many to one mapping. Therefore, the weight value and model structure can be updated simultaneously during the online incremental learning process. To testify the effectiveness of the presented method, experiments of tool condition classification in the process of end milling of Titanium alloy are carried out and two incremental learning cases are simulated. The analysis of online learning process in both cases shows that the structure and parameters of the model can be adjusted automatically without requiring access to the previous training data. At the same time, the accuracy analysis demonstrates that the presented method has strong ability to learn the new knowledge without forgetting the previous knowledge.
机译：刀具磨损的状态监测对于保证工件质量和延长刀具寿命至关重要。为了提高训练速度和增量学习的灵活性，开发了一种改进的Fuzzy ARTMAP分类器，其中共振层与类别节点通过多对一映射直接链接。因此，可以在在线增量学习过程中同时更新权重值和模型结构。为了验证该方法的有效性，进行了钛合金立铣削过程中刀具状态分类的实验，并模拟了两个增量学习案例。对这两种情况的在线学习过程的分析表明，该模型的结构和参数可以自动调整，而无需访问以前的训练数据。同时，精度分析表明，该方法具有较强的学习新知识的能力，并且不会忘记以前的知识。

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