基于深度学习和多传感器的数控机床铣刀磨损状态信号监测方法研究

由于单一传感器存在获取信息量有限、抗干扰能力较弱等问题及传统网络模型诊断时间长、诊断率低等现象,采用振动、噪声等多个传感器监测铣刀的磨损状态。提出将深度学习和多传感器相结合的铣刀磨损状态信号监测方法;将经核主元筛选和未筛选的数据分别输入到BP神经网络、RBF神经网络和深度卷积神经网络中进行模式识别,并对识别结果进行对比和分析。结果表明:深度学习和多传感器相结合的铣刀磨损状态监测方法在特征量比较大、数据量比较多的情况下诊断速度、准确率均比较高,在铣刀磨损状态监测中具有明显的优势。

Milling Cutter Wear Condition Signal Monitoring Method of CNC Based on Deep Learning and Multi-sensor

Due to the limited information,weak anti-interference ability of a single sensor,and the long diagnosis time and low diagnosis rate of traditional network model,several sensors such as vibration and noise sensors were used to monitor the wear state of the milling cutter.A milling cutter wear condition signal monitoring method by combining deep learning and multi-sensor was proposed.The pattern recognition was carried out by inputting the screened and unscreened data into BP neural network,RBF neural network and deep convolutional neural network respectively,and the recognition results were compared and analyzed.The results show that the wear condition monitoring method combined with deep learning and multi-sensor has high diagnostic speed and accuracy when the characteristic quantity and the data quantity are large,and has obvious advantages in monitoring the wear condition of milling cutter.