| 李倩*,刘惠康*,皮瑶*,喻青**.基于深度置信网络的制动器故障诊断方法[J].高技术通讯(中文),2021,31(10):1075~1080 |
| 基于深度置信网络的制动器故障诊断方法 |
| Brake fault diagnosis method based on deep belief network |
| |
| DOI：10.3772/j.issn.10020470.2021.10.008 |
| 中文关键词: 深度置信网络（DBN）； 吊车制动器； 故障诊断； 柔性薄膜传感器阵列 |
| 英文关键词: deep belief network (DBN), brake of crane，fault diagnosis, flexible thin film sensor array |
| 基金项目: |
|
| 摘要点击次数: 1976 |
| 全文下载次数: 1283 |
| 中文摘要: |
| 针对吊车制动器故障诊断中故障机理复杂、先验知识欠缺、传统的故障诊断方法精度不高和人工依赖大等问题，本文提出一种基于深度置信网络（DBN）的制动器故障诊断方法。该方法通过柔性薄膜传感器阵列获取制动器不同工况的实时特征数据及信号，利用网络的无监督特征学习和有监督微调，构建制动器故障诊断的深层网络模型，从而实现了对制动器的故障诊断及预测。最后，分别与支持向量机（SVM）和遗传算法（GA）优化的BP神经网络（GA-BP）进行了对比研究，通过实验证明了本文方法的优越性。 |
| 英文摘要: |
| In order to solve the problems of complex fault mechanism, lack of prior knowledge, low precision and artificial dependence of traditional brake fault diagnosis methods, a fault diagnosis method based on deep belief network (DBN) is proposed. In this method, the real-time characteristic data and signals of brake under different working conditions are obtained by flexible thin film sensor array, and the deep network model of brake fault is constructed by using unsupervised feature learning and supervised fine-tuning of training network, so as to realize the fault diagnosis and prediction of the brake. Finally, the diagnosis results are compared with those of support vector machine (SVM) and BP neural network optimized by genetic algorithm (GA), which proved the superiority of this method by experiments. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
