基于DTCNN-SVM的工业循环水系统供水泵故障诊断

Fault diagnosis of water supply pump in industrial circulating water system based on DTCNN-SVM

工业循环水系统供水泵的工作状态是影响工业过程安全生产的重要因素,为及时准确地识别供水泵的工作状态,提出一种基于深度迁移卷积神经网络和支持向量机(deep transfer convolutional neural network-support vector machine,DTCNN-SVM)的故障诊断方法。将与工作状态强相关的振动信号进行信号-图像预处理,实现振动时序信号的二维灰度图化;在此基础上,采用融合迁移学习与残差神经网络的深度迁移卷积神经网络模型提取振动信号灰度图特征,并基于模糊不一致性度量对深度学习特征进行约简;采用支持向量机法建立供水泵故障诊断模型。试验结果表明,所提方法在少量样本数据和模型参数下能有效识别供水泵工作状态。

Working state of water supply pump in industrial circulating water system is an important factor affecting safe production of industrial process.Here,to identify working state of water supply pump timely and accurately,a fault diagnosis method based on deep transfer convolution neural network and support vector machine(DTCNN-SVM)was proposed.Firstly,signal-image pre-processing of vibration signals strongly correlated to working state was performed to obtain two-dimensional gray-scale images of vibration time sequence signals.Then,DTCNN model fusing transfer learning and residual neural network was used to extract features of vibration signal gray-scale images,and deep learning features were reduced based on fuzzy inconsistency measure.Finally,SVM was used to establish the fault diagnosis model of water supply pump.The test results showed that the proposed method can effectively identify working state of water supply pump under the condition of a small amount of sample data and model parameters.