基于空洞卷积神经网络的旋转机械故障诊断方法

由于旋转机械的振动信号具有非平稳、复杂多样、数据量大的特点,传统的方法难以较好地实现旋转机械故障诊断。近年来,基于深度学习的故障诊断算法发展迅速,其中,卷积神经网络(Convolutional Neural Network,CNN)由于可实现自动提取特征、运算效率高等优点受到广泛关注,但在识别准确率等方面仍然存在部分问题。为实现多传感器监测状态下的旋转机械故障诊断,在经典卷积神经网络结构的基础上,引入了多通道数据融合处理、空洞卷积层、批标准化处理、PReLU激活函数、全局平均池化层等改进方法,构造了一种新型的、高效的空洞卷积神经网络(Atrous Convolution-Convolutional Neural Network,AC-CNN),并基于该模型进行了旋转机械故障诊断实验。实验结果表明,提出的故障诊断模型分类准确率可达99%以上,对比其他神经网络方法具有明显优势。

Fault Diagnosis of Rotating Machinery Based on Atrous Convolution-Convolutional Neural Network

Because the vibration signals of rotating machinery are unstable,complex and large,the traditional methods are not suitable for its fault diagnosis.In recent years,fault diagnosis algorithms based on deep learning have developed rapidly.Among them,convolutional neural network (CNN) has attracted wide attention due to automatic feature extraction and high efficiency.However,there are still many problems of applying CNN directly to fault diagnosis.For multi-sensor monitoring of rotating machinery fault diagnosis,based on the general CNN,a novel and efficient Atrous Convolution-Convolutional Neural Network (AC-CNN) is constructed by introducing the data fusion,Atrous Convolutional,Batch Normalization,PReLU,and Global Average Pooling.Based on AC-CNN,fault identification of rotating machinery vibration signal is realized.The experimental results show that the classification accuracy of the proposed fault diagnosis model can reach more than 99%,which has obvious advantages compared with other neural networks.