论文部分内容阅读
为了实现刀具磨损状态的自动识别 ,采用机床功率法进行了刀具自然磨损和不同切削参数 (切削速度、进给量和切削深度 )对功率信号影响的实验 .在此基础上 ,建立了功率信号的时序AR模型 .在提取作为刀具磨损特征量的 AR模型参数时 ,考虑了切削用量对模型参数的影响 ,提出了特征量选取的准则 ,使所提取的特征量更加实用化 .通过具有自学习和良好函数逼近能力的神经网络获得了特征量对刀具状态的隶属函数 ,并利用模糊神经网 Fuzzy ART实现了刀具磨损状态的自动识别 ,识别正确率为 95 % ,说明所提出的方法是有效可行的
In order to realize the automatic identification of tool wear state, experiments on the influence of natural wear of tool and different cutting parameters (cutting speed, feed rate and cutting depth) on power signal were carried out by machine tool power method.On the basis of this, the power signal The time series AR model.When extracting the parameters of the AR model which is the tool wear characteristic quantity, the influence of the cutting quantity on the model parameters is considered, and the criterion of the selection of the characteristic quantity is proposed so that the extracted characteristic quantity is more practical.Through the self- The neural network with good function approximation ability obtains the membership function of the feature quantity to the tool state, and uses the fuzzy neural network to realize the automatic identification of the tool wear state. The correct recognition rate is 95%, which shows that the proposed method is effective and feasible