基于深度卷积网络的多工况寿命预测方法研究

机电设备的寿命预测是状态维修中的一项重要任务,目前在多工况条件下的机电设备寿命预测效果并不理想,为了更好的预测多工况条件下的设备剩余寿命。文章对现有的涡轮风扇发动机开源数据集进行了研究,提出了一种新的多工况深度卷积神经网络模型(MC-DCNN)来估计剩余寿命。将原始数据输入文章提出的MC-DCNN模型中,模型输出不同工况下的设备剩余寿命。该模型能更好的预测多工况设备的剩余寿命,在实际生产中也更有价值。最后通过对公开数据集进行实验,并与现有的模型进行分析对比,证明该模型的有效性。

Multi-condition Remaining Useful Life Prediction Method Based on Deep Convolution Network

Remaining useful life prediction of mechanical and electrical equipment is an important task in condition-based maintenance. At present, the effect of remaining useful life prediction of mechanical and electrical equipment under multi-condition is not ideal. In order to predict the residual life of equipment under multi-condition better. In this paper, the existing open source data sets of turbofan engines are studied.,and a new multi-condition deep convolution neural network model(MC-DCNN) is proposed to estimate the residual life. Input the original data into the MC-DCNN model proposed in this paper, the model outputs the residual life of the equipment under different operating conditions. The model can better predict the residual life of multi-condition equipment and is more valuable in actual production. Finally, the effectiveness of the model is proved by experiments on open data sets and comparisons with existing models.