A Directed Acyclic Graph Network Combined With CNN and LSTM for Remaining Useful Life Prediction

Jialin Li,David He,Xueyi Li

doi:10.1109/access.2019.2919566

Abstract

Accurate and timely prediction of remaining useful life (RUL) of a machine enables the machine to have an appropriate operation and maintenance decision. Data-driven RUL prediction methods are more attractive to researchers because they can be deployed quicker and cheaper compared to other approaches. The existing deep neural network (DNN) models proposed for the applications of RUL prediction are mostly single-path and top-down propagation. In order to improve the prognostic accuracy of the network, this paper proposes a directed acyclic graph (DAG) network that combines long short term memory (LSTM) and a convolutional neural network (CNN) to predict the RUL. Different from the existing prediction models combined with CNN and LSTM, the method proposed in this paper combines CNN and LSTM organically instead of just using CNN for feature extraction. Moreover, when a single timestamp is used as an input, padding the signals in the same training batch would affect the prediction ability of the developed model. To overcome this drawback, the proposed method generates a short-term sequence by sliding the time window (TW) with one step size. In addition, based on the degradation mechanism, the piece-wise RUL function is used instead of the traditional linear function. In the experimental test, the turbofan engine degradation simulation dataset provided by NASA is used to validate the proposed RUL prediction model. By comparing with the existing methods using the same dataset, it can be concluded that the prediction method proposed in this paper has better prediction capability.

Highlights

The prognostics and health management (PHM) of the mechanical equipment has received much attention, and the prediction of remaining useful life (RUL) is the core of the PHM [1], [2]
This paper proposes a directed acyclic graph (DAG) network based on long short term memory (LSTM) and convolutional neural network (CNN) to improve the accuracy of the RUL prediction of the machines
Different from the DAG structure presented in this paper, the existing prediction models of using both LSTM and CNN are all combined in a serial manner

Summary

Introduction

The prognostics and health management (PHM) of the mechanical equipment has received much attention, and the prediction of remaining useful life (RUL) is the core of the PHM [1], [2]. There are three types of methods for predicting the RUL of mechanical equipment: model-based prognostics, data-driven prognostics, and hybrid approaches. Model-based prediction uses a physical understanding (physical model) of the system to predict the RUL. It can be further divided into microlevel models [4] and macro-level models [5] based on the. A macro-level model is a simplified representation of the system. It defines the relationship between input variables, state variables, and system output. Modeling strategies for data-driven prediction methods can be of two types: 1) modeling cumulative damage and inferring the damage threshold, 2) learning the RUL directly from the data

Methods

Results

Conclusion