基于深度神经网络的旋转机械故障预测模型的研究
Research on Fault Prediction Model of Rotating Machinery Based on Deep Neural Network

本文针对化工驱动旋转机械设备-风机的故障，提出基于深度置信网络的故障预测模型，通过数据采集系统的设计、数据集的构建、特征提取、深度神经网络的模型训练、预测结果输出五个步骤来进行，这一模型的优点为无需前期进行知识模型的构建，是基于数据的信息获取及模型构建，对于工业设备的故障预测具有良好的效果。
Oriented chemical driven rotating machinery equipment-fan fault, the paper proposes a kind of fault prediction model based on deep belief network, through the design of data acquisition system, the construction of a data set, feature extraction, the depth of the neural network model training, prediction results output five steps. The advantage of this model is without prior to build knowledge model, which is based on the data of information acquisition and model building, and the industrial equipment failure prediction has a good effect.