摘要
为了实现同时辨别不同种类的切削液并对其质量进行分析,在数据预处理基础上,提出一种基于向量夹角余弦法和神经网络的切削液辨别分析方法。首先,鉴于切削液的组分复杂,对其组分数据进行标准化及主成分降维处理;其次,利用夹角余弦法与神经网络分析得到未知样品与标准指纹图谱之间识别相似度并对比辨别切削液种类,并采用阈值算法对两种方法测得的结果进行一致性验证;最后,结合实际应用验证了该方法的科学性和可靠性,从而为切削液辨别分析提供有效的理论参考。
To identify different types of cutting fluids at the same time and analyze their quality,based on the data preprocessing,a cutting fluid identification analysis method based on vector angle cosine method and neural network is proposed.First,because the components of the cutting fluid are complex,the component data is standardized and the principal component dimension reduction processing is performed.Then use the angle cosine method and neural network analysis to obtain the similarity between the unknown sample and the standard fingerprint,compare and identify the type of cutting fluid,and use the threshold algorithm to verify the consistency of the results measured by the two methods.Finally,the scientificity and reliability of the method were verified in combination with practical applications,so as to provide an effective theoretical reference for discriminating analysis of cutting fluids.
作者
李波
汪永超
李强
吴爱强
LI Bo;WANG Yong-chao;LI Qiang;WU Ai-qiang(School of Mechanical Engineering,Sichuan University,Chengdu 610065,China)
出处
《组合机床与自动化加工技术》
北大核心
2020年第10期76-79,84,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金资助项目(51875370)。
关键词
切削液
模式识别
向量夹角余弦法
神经网络
阈值算法
cutting fluid
pattern recognition
vector angle cosine method
neural network
threshold algorithm
