摘要
Abstract
Due to the particularity of special equipment,fault detection is particularly important,and detecting vibration signals of special equipment is an important means to detect faults.For this reason,this paper proposes a method for detecting vibration signals of special equipment based on sparse adaptive S-transform.Starting from the time-frequency characteristics of the vibration signal of special equipment,it extracts the time-frequency characteristic map of the vibration signal of special equipment through sparse adaptive S-transform,constructs a deep convolutional neural network model,and uses the time-frequency feature map extracted through sparse adaptive S-transform as the input sample of the network model.After in-depth learning,it completes the detection of special equipment fault vibration signals,and obtains equipment fault diagnosis results.The experimental results show that the method can extract better vibration signal features,clearly express the fault frequency,and have strong feature expression ability.It can clearly detect the occurrence time and cause of special equipment faults with a high detection accuracy.关键词
特种设备/稀疏自适应S变换/时频特征/振动信号/故障检测/深度卷积神经网络Key words
special equipment/sparse adaptive S-transform/time frequency characteristics/vibration signal/fault detection/deep convolution neural network
