基于朴素贝叶斯的有效样本提取技术

An Effective Sample Extraction Technique Based on Naive Bayes

在数控加工过程中会产生海量数据,如数控系统电控数据、工艺任务数据、切削响应数据等,其中存在数控加工中工况信息和响应信息自动标记困难,而且存在大量冗余、无效信息,当前缺乏有效样本信息提取算法。针对上述问题,文中研究了基于指令域的铣削加工工艺数据的标记技术,然后提出一种从标记数据中提取有效样本的算法,最后通过有效样本的提取与验证试验证实了该提取算法的有效性,为后续智能模型的训练奠定了坚实的基础。

In the process of NC machining,massive data will be generated,such as electronic control data of NC system,process task data,cutting response data,etc.However,on the one hand,it is difficult to automatically mark working condition information and response information in NC machining.On the other hand,there are a lot of redundant and invalid information,and there is a lack of effective sample information extraction algorithm.In order to solve the above problems,this paper studies the marking technology of milling process data based on instruction domain,and then proposes an algorithm to extract effective samples from the marking data.Finally,through the extraction and verification experiments of effective samples,the effectiveness of the extraction algorithm in this paper is verified,which lays a solid foundation for the subsequent training of intelligent model.