Fuel cell (FC) is a promising alternative energy source in a wide range of applications. Due to the unsatisfactory durability performance, FC has not yet been widely used. Prognostics and health management (PHM) has been demonstrated to be an effective solution to enhance the FC durability performance by predicting FC degradation characteristics and adopting health condition based control and maintenance. As the primary task of PHM, prognostics seeks to estimate the remaining useful life (RUL) of FC as early and accurately as possible. However, when FC faces dynamic operating conditions, its degradation characteristics are often hidden in the complex system dynamic behaviors, which makes prognostics challenging. To address this issue, a hybrid prognostics approach is proposed in this paper. Specifically, the health indicator of FC is extracted using a degradation behavior model and sliding-window model identification method. Subsequently, a symbolic-based long short-term memory networks (LSTM) is used to predict the health indicator degradation trend and estimate the RUL. The experimental and simulation results show that the proposed model is able to describe the dynamic behavior of the FC stack voltage and the extracted health indicator show a significant degradation trend. Moreover, health indicator prediction and RUL estimation performance can be improved by deploying the proposed symbolic-based LSTM prognostics model. The proposed approach provides a prognostic horizon approaching 50% of the FC life-cycle, and the average relative accuracy of estimated RUL is close to 90%.
The prognostic of proton exchange membrane fuel cells (PEMFCs) degradation and the estimation of its remaining useful life (RUL) are effective ways to improve the reliability of the target system and reduce maintenance costs, which is of great significance for the wide commercialization of PEMFCs. Many factors cause the degradation of PEMFCs, and these factors are often difficult to measure accurately. The prognostic method based on long short-term memory networks (LSTMs) has better memory ability for time series and has been demonstrated able to describe the degradation trend of PEMFCs. However, the traditional LSTM prediction algorithm seems to easily fall into the local optimal solution in long-term prediction cases. Overfitting like errors may result in an imprecise or even unstable prognostic. This paper proposes a novel method, named navigation sequence driven LSTMs (NSD-LSTMs), to enhance the accuracy of PEMFCs degradation trend prediction. Two types of PEMFCs aging test data under different load conditions were used to verify the performance of NSD-LSTMs. Experimental results show that, compared with traditional LSTMs, NSD-LSTMs can improve the accuracy of trend prediction. Accurate degradation prognostic can be used to predict RUL and provide guidance for the commercial application of PEMFCs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.