Steady-State Process Fault Detection for Liquid Rocket Engines Based on Convolutional Auto-Encoder and One-Class Support Vector Machine

Xiaobin Zhu,Xing Cui,Jianjun Wu,Runsheng Hu,Yuqiang Cheng

doi:10.1109/access.2019.2961742

Abstract

Liquid rocket engines (LREs) are the main propulsive devices of launch vehicles. Due to the complex structures and extreme working environments, LREs are also the components prone to failure. It is of great engineering significance to develop fault detection technologies which can detect fault symptoms in time and provide criteria for further fault diagnosis and control measures to avoid serious consequences during both the ground tests and flight missions. This paper presents a novel fault detection method based on convolutional auto-encoder (CAE) and one-class support vector machine (OCSVM) for the steady-state process of LREs. We train the CAEs by normal ground hot-fire test data of a certain type of large LRE for automatic feature extraction. Then the obtained features are used to train the OCSVMs to accomplish the fault detection task. The results demonstrate that the proposed method outperforms traditional redline system (RS), adaptive threshold algorithm (ATA), and back-propagation neural network (BPNN). We also study the effect of sample sizes and domain knowledge on the performance of the proposed method. The results suggest that appropriate measures that enrich the effective information content in the training data, such as increasing sample size and introducing domain knowledge, can further improve the performance of the proposed fault detection method.

Highlights

As an irreplaceable propulsive device, the liquid rocket engine (LRE) is one of the most critical components of the launch vehicle
FAULT DETECTION METHOD FOR STEADY-STATE PROCESS OF LIQUID ROCKET ENGINE we develop a fault detection method consisting of a convolutional auto-encoder (CAE) based feature extraction module and an one-class support vector machine (OCSVM) based fault detection module for the LRE steady-state process
We propose a fault detection method based on CAE and OCSVM for the steady-state process of a certain type of large LRE

Summary

INTRODUCTION

As an irreplaceable propulsive device, the liquid rocket engine (LRE) is one of the most critical components of the launch vehicle. X. Zhu et al.: Steady-State Process Fault Detection for LREs Based on CAE and OCSVM attention and been widely studied in many fields, such as science, business, and government [13]. Sarmiento et al [32] proposed a method based on OCSVM and principal component analysis (PCA) to detect faults in reactive ion etching systems through optical emission spectroscopy data. The fault detection process is usually assisted by certain feature extraction or selection methods, such as PCA [32] and WTFE [35]. We propose a fault detection method for the steady-state process of LRE based on convolutional auto-encoder (CAE) and OCSVM.

PROBLEM BACKGROUND

ONE-CLASS SUPPORT VECTOR MACHINE

OUTLINE OF FAULT DETECTION METHOD

RESULTS AND DISCUSSIONS

CONCLUSION AND FUTURE WORKS