Automated Machine Learning for Remaining Useful Life Estimation of Aircraft Engines

Abstract

Remaining useful life (RUL) of an asset or system is defined as the length from the current time and operating state to the end of the useful life. It is of paramount importance for safety-critical industries such as aviation and lies in the heart of prognostics and health management (PHM). This paper investigates the usage of automated machine learning (AutoML) for RUL estimation, based on using classical machine learning algorithms for regression. The data is pre-processed by extracting statistical features from expanding windows of the signal in order to uncover the degradation that has been accumulating from the early life of the system or after an overhaul. We evaluate our methodology on the widely-used C-MAPSS dataset and compare our approach to the state-of-the-art deep neural networks (DNNs) and classical machine learning algorithms. The experimental results show that AutoML outperforms or is comparable to traditional machine learning techniques and standard neural networks, while being outperformed by specifically designed neural networks on datasets with multiple fault mode and operating conditions. These results show that with the correct pre-processing automated machine learning is able to accurately estimate the RUL, which implies that such approaches can be industrially deployed.

End-to-End Pipeline for Uncertainty Quantification and Remaining Useful Life Estimation: An Application on Aircraft Engines

“…As mentioned in Section 1, amongst the data-driven methods employed for prognostics DNNs have proven to be good candidates due to their representational power (Lei et al, 2018;Benker et al, 2021;Kefalas et al, 2021;Caceres et al, 2021;Peng et al, 2020;B. Wang et al, 2020).…”

“…The remaining useful life is not merely a target variable that can be predicted from sensor measurements, but it is a variable that needs to be inferred from a longer trend of degradation patterns and when those begin to occur. In this view, and due to the advances in the general field of artificial intelligence (AI), deep learning (DL) and DNNs have proven to be a successful candidate to the RUL estimation task (Lei et al, 2018;Benker, Furtner, Semm, & Zaeh, 2021;Kefalas, Baratchi, Apostolidis, van den Herik, & Bäck, 2021;Caceres, Gonzalez, Zhou, & Droguett, 2021;Peng, Ye, & Chen, 2020;B. Wang, Lei, Yan, Li, & Guo, 2020).…”

Life Prediction under Charging Process of Lithium-Ion Batteries Based on AutoML

“…However, existing AutoML research on RUL prediction is just beginning and challenges do emerge, as dealing with the long multivariate time-series problem requires extensive data pre-processing and feasible feature extraction, to ensure that useful information can be accumulated and transmitted. All the three published articles [21][22][23] proposed RUL prediction of aircraft engines based on AutoML using a simulated turbofan engine degradation open-source dataset from NASA PCoE [24]. Kefalas et al [21] used a mature architecture TPOT [25] in automatic modeling to develop and optimize machine learning pipelines in an automatic manner, introducing expanding windows to extract statistical features to evaluate the degradation accumulated in the early life of the system.…”

“…All the three published articles [21][22][23] proposed RUL prediction of aircraft engines based on AutoML using a simulated turbofan engine degradation open-source dataset from NASA PCoE [24]. Kefalas et al [21] used a mature architecture TPOT [25] in automatic modeling to develop and optimize machine learning pipelines in an automatic manner, introducing expanding windows to extract statistical features to evaluate the degradation accumulated in the early life of the system. Tornede et al [22] pointed out a cooperative coevolutionary algorithm, which enlarges the number of pipelines that are explored in a single run, through coevolving the two populations, which are in sub-spaces partitioned by search space into feature extraction and regression methods.…”

An Efficient Contesting Procedure for AutoML Optimization

“…Machine learning (ML) has achieved significant advances in a variety of real-world applications [1]- [3]. To achieve this goal, a practitioner needs to choose a well-performing sequence of algorithms, a.k.a, the ML pipeline, for the given problem and tune its hyperparameters to maximize the performance.…”