基于柔性薄膜阵列压力传感器的抱闸故障诊断  

Fault diagnosis of brake by flexible film array pressure sensor

作　　者：皮瑶 刘惠康[1] 李倩[1] Pi Yao;Liu Huikang;Li Qian(Engineering Research Center of Metallurgical Automation and Measurement Technology,Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081)

机构地区：[1]武汉科技大学教育部冶金自动化与检测技术工程研究中心,武汉430081

出　　处：《高技术通讯》2021年第8期836-843,共8页Chinese High Technology Letters

基　　金：国家重点研发计划(2017YFC0805104)资助项目。

摘　　要：抱闸制动装置广泛应用于工业提升装置、民用曳引式电梯,针对抱闸制动器的运行状态监测和故障诊断,本文提出一种基于柔性薄膜阵列压力传感器的故障诊断方法,运用卷积神经网络(CNN)对传感器的数据进行处理达到故障诊断的目的。本文在LeNet5模型的基础上引入跨连接部分,将网络结构中提取的低层次特征与高层次特征相结合,经过全连接层达到多分类的目的。通过训练来自柔性薄膜阵列压力传感器的实验数据,该模型实现了4种基本抱闸故障和正常状态的自动识别。实验结果表明,改进的LeNet卷积神经网络模型在抱闸故障诊断上的检测正确率达到99.19%,该模型在同一训练数据集上的表现明显优于传统的LeNet5模型。Brake device is widely used in industrial and civil traction elevator.Aiming at the condition monitoring and fault diagnosis of brake,a method based on flexible film array pressure sensor is proposed in this paper.Convolution neural network(CNN) is used to process the sensor data to achieve the purpose of fault diagnosis.This paper adds the cross-connected part which combines the low-level features with the high-level features to the classical LeNet 5 convolution neural network,and achieves the goal of multi classification through the full connection layer.This method can identify four basic faults and normal condition automatically by training the experimental data of flexible film array pressure sensor.The experimental results show that the detection accuracy of the improved LeNet convolution neural network reaches 99.19%,and the performance of this model is better than that of the traditional LeNet 5 model on the same training data set.

关 键 词：柔性薄膜阵列压力传感器 LeNet模型 跨连接 抱闸制动器 故障诊断 

分 类 号：TH17[机械工程—机械制造及自动化] TP212[自动化与计算机技术—检测技术与自动化装置]