ArticlePDF Available

Dynamic Semiempirical PEMFC Model for Prognostics and Fault Diagnosis

Authors:

Abstract and Figures

This paper introduces a dynamic semiempirical model that predicts the degradation of a proton exchange membrane fuel cell (PEMFC) by introducing time-based terms in the model. The concentration voltage drop is calculated using a new statistical equation based on the load current and working time, whereas the ohmic and activation voltage drops are updated using time-based equations borrowed from the existing literature. Furthermore, the developed model calculates the membrane water content in the PEMFC, which indicates the membrane hydration state and indirectly diagnoses the flooding and drying faults. Moreover, the model parameters are optimized using a recently developed butterfly optimization algorithm. The model is simple and has a short runtime; therefore, it is suitable for monitoring. Voltage degradation under various loading currents was observed for long working hours. The obtained results indicate a significant degradation in PEMFC performance. Therefore, the proposed model is also useful for prognostics and fault diagnosis.
Content may be subject to copyright.
A preview of the PDF is not available
... 0 : The effective exchange current is a function of the electrode catalyst loading and the catalyst specific surface area (Barbir, 2013). For the fuel cell operated under dynamic load, the cycling will lead to the major degradation of the electrodes: the catalyst layer degradation and the carbon support degradation, especially, the catalyst loss is aggravated by the potential cycles (Khan, et al., 2021). : The limiting current on the cathode varies due to the changes on the diffusivity of oxygen, the gas pressure and the thickness of the gas diffusion layer (Morgan & Datta, 2014). ...
Article
Full-text available
Proton exchange membrane (PEM) fuel cell has seen its recent increasing deployment in both automotive and stationary applications. However, the unsatisfied durability of the fuel cell has barriered in the way of its successful commercialization. Recent research on prognostics and predictive maintenance has demonstrated its effectiveness in predicting the system failure and improving the durability of the PEM fuel cell. This paper contributes to developing a degradation identification method for the PEM fuel cell operating under dynamic load. A degradation indicator is proposed based on the polarization model and the nonlinear regression method is applied to extract the degradation feature by segmenting the voltage measurement. To perform prognostics, a machine learning method based on a multi-step echo state network is developed, in which a sliding window is used to recursively reformulate the input sequence with predicted values in the prediction phase. The length of the sliding window is optimized by a genetic algorithm. The proposed method is verified on the experimental PEM fuel cell degradation data and improves the prediction performance on both accuracy and computation speed when comparing with other prognostics methods.
Article
Proton exchange membrane fuel cell (PEMFC) is a promising vehicle power because of its high energy conversion efficiency and pollution-free reactant. However, the complex structure of multi-system coordination makes the correlation between parameters strong, the fault rate is high and the control is difficult. Timely and accurate fault diagnosis and effective fault-tolerant control are of great significance to the stable output and durability of PEMFC system. Firstly, this paper classifies different faults and describes the applicable scenarios of each classification method. Then, the fault diagnosis methods based on experiment, model and data are summarized, and their advantages and disadvantages are compared. Then, the methods and characteristics of hardware fault tolerance and software fault tolerance are analyzed from the point of redundancy, and the application of fault tolerance control in maintaining the stability of PEMFC system is summarized. Finally, the improvement direction and development prospect of fault diagnosis related technology of vehicle PEMFC system are proposed.
Article
Full-text available
Proton exchange membrane fuel cell PEMFC is nonlinear source so modeling its output is essential. Also PEMFC is extremely prone to change in ambient conditions. Previously the complex modeling techniques have been used to model PEMFC voltage, but these technique are not good for online purposes. Hence a semiempirical approach has been suggested for PEMFC voltage prediction which uses the combination of theoretical and empirical equations. But most of the semiempirical models have lot of deficiencies which need to be corrected with the help of experimentation of different PEMFC systems at different conditions. In this research paper, a novel semiempirical model has been chosen which is less complex, dynamic and has the ability to diagnose fault as well. The model is only tested on two PEMFC stacks but never tested on single PEMFC. In this research, the model has been tested on single-cell PEMFC system and the parameters are optimized by using lightening search algorithm. The temperature and voltage model have been validated, and the new optimized parameters are recorded for different ambient conditions. The model discrepancies have been identified, and the new equations for parameters have been proposed which can be helpful in the making of generic model.
Article
Full-text available
This paper presents a dynamic temperature model for a proton exchange membrane fuel cell (PEMFC) system. The proposed model overcomes the complexity of conventional models using first-order expressions consisting of load current and ambient temperature. The proposed model also incorporates a PEMFC cooling system, which depends upon the temperature difference between events. A dynamic algorithm is developed to detect load changing events and calculate instantaneous PEMFC temperature variations. The parameters of the model are extracted by employing the lightning search algorithm (LSA). The temperature characteristics of the NEXA 1.2 kW PEMFC system are experimentally studied to validate model performance. The results show that the proposed model output and the temperature data obtained from experiments for linear and abrupt changes in PEMFC load current are in agreement. The root-mean-square error between the model output and experimental results is less than 0.9. Moreover, the proposed model outperforms the conventional models and provides advantages such as simplicity and adaptability for low and high sampling data rates of input variables, namely, load current and ambient temperature. The model is not only helpful for simulations but also suitable for dynamic real-time controllers and emulators.
Conference Paper
Full-text available
Proton Exchange Membrane Fuel Cells are promising energy converters that allow to power vehicles or buildings in a clean manner. Nevertheless, their performance are affected by faults and irreversible degradation mechanisms that are far from being fully understood. Consequently, during the last decade, researches have been conducted on the diagnostic of faults of this promising converter. Nevertheless, aging was never the subject of a particular attention concerning control. As a result, this paper proposes an aging tolerant control strategy for Proton Exchange Membrane Fuel Cells. It aims at generating the load current reference taking the state of health into account. Moreover, using a model inversion of an Energetic Macroscopic Representation with time-varying parameters, the coherent references of input flows of gas can be calculated. Finally, the paper details a Maximum Power Point Tracking algorithm which allows to identify the maximum power the fuel cell is able to provide at each time.
Article
Monitoring the transmission status of multi-joint industrial robots is very important for the accuracy of the robot motion. The fault diagnosis information is an indispensable basis for the collaborative maintenance of the robots in industry 4.0. In this paper, an attitude data-based intelligent fault diagnosis approach is proposed for multi-joint industrial robots. Based on the analysis of the transmission mechanism, the attitude change of the last joint is employed to reflect the transmission fault of robot components. An economical data acquisition strategy is performed by only installing one attitude sensor on the last joint of the multi-joint robot. Considering the characteristics of attitude data, a hybrid sparse auto-encoder (SAE) and support vector machine (SVM) approach, namely SAE-SVM, is subsequently presented to construct an intelligent fault diagnosis model by learning from the attitude dataset with multiple fault information. Experimental results show that the proposed fault diagnosis approach has promising performance in identifying different faults related to the reducer of a 6-axial multi-joint industrial robot accurately and reliably.
Article
This paper proposes a new hybrid degradation model of Proton Exchange Membrane Fuel Cell (PEMFC) used in Fuel Cell Electric Vehicle (FCEV) operating under real conditions. This hybrid method is based on the wavelet analysis, Extreme Learning Machine (ELM) and Genetic Algorithm (GA). This new degradation model considers the influence of PEMFC load current, relative humidity, temperature, and hydrogen pressure. First, the wavelet analysis is used to decompose the voltage aging waveform of PEMFC into multiple sub-waveforms. Second, the ELM is applied to build the degradation model of each sub-waveform. Then, the overall degradation model of PEMFC is obtained by combining the degradation model of each sub-waveform. Finally, GA is adopted to improve the global optimization of the proposed degradation model. This model is experimentally validated by using actual data from 3 PEMFCs in 3 FCEVs carrying out postal mail delivery missions in the field. The results show that the proposed hybrid method can get higher accuracy than several traditional approaches.
Article
Echo state network (ESN) is a fast recurrent neural network with remarkable generalization performance for intelligent diagnosis of machinery faults. When dealing with high dimensional signals mixed with much noise, however, the performance of a deep ESN is still highly affected by the random selection of input weights and reservoir weights, resulting in the optimal design of the deep ESN architecture an open issue. For this reason, a hybrid evolutionary algorithm featuring a competitive swarm optimizer combined with a local search is proposed in this paper. An indirect encoding method is designed based on the network characteristics of ESN to make the evolutionary process computationally economical. A layer-wise optimization strategy is subsequently introduced for evolving deep ESNs. The results of two experimental cases show that the proposed approach has promising performance in identifying different faults reliably and accurately by comparing with other intelligent fault diagnosis approaches.
Article
Proton exchange membrane fuel cell (PEMFC) models are multivariate with different nonlinear elements which should be identified accurately to assure dependable modeling. Metaheuristic algorithms are perfect candidates for this purpose since they do an informed search for finding the parameters. This paper utilizes three algorithms, namely shuffled frog-leaping algorithm (SFLA), firefly optimization algorithm (FOA), and imperialist competitive algorithm (ICA) for the PEMFC model calibration. In this regard, firstly, the algorithms are employed to find the parameters of a benchmark PEMFC model by minimizing the sum of squared errors (SSE) between the measured and estimated voltage for two available case studies in the literature. After conducting 100 independent runs, the algorithms are compared in terms of the best and the worst SSEs, the variance, and standard deviation. This comparison indicates that SFLA marginally outperforms ICA and FOA regarding the best SSE in both cases while it performs 20% and twofold better than other algorithms concerning the worst SSE. Furthermore, the obtained variance and standard deviation by SFLA are much less than the other algorithms showing the precision and repeatability of this method. Finally, SFLA is used to calibrate the model for a new case study (Horizon 500-W PEMFC) with variable temperature.
Article
Proton exchange membrane fuel cell (PEMFC) degradation prediction is essential especially in transportation applications, since one of the major issues that hinder its worldwide commercialization is represented by its durability. However, due to the complex physical phenomena inside the fuel cell, which are usually strongly inter-coupled, the conventional semi-empirical model-based prognostics approach may fail to predict the aging phenomena under varies fuel cell operating conditions. In order to improve prognostics accuracy, this paper proposed a data-fusion approach to forecast the fuel cell performance based on long short-term memory (LSTM) recurrent neuron network (RNN) and auto-regressive integrated moving average (ARIMA) method. LSTM can efficiently make a prediction regarding a long-term physical degradation, whereas the fusion with ARIMA can effectively track the degradation tendency. In order to validate the performance of the proposed data-fusion approach, two different PEMFCs are tested for recording the aging experimental datasets. The forecasting results indicate that the proposed LSTM-ARIMA approach can accurately predict PEMFC degradation, which can be then used directly to optimize fuel cell performance implemented in transportation applications.
Article
Proton exchange membrane fuel cell is an emerging renewable energy resource for transportation and power generation. Similar to other renewable resources, the performance of proton exchange membrane fuel cell is affected by ambient conditions. However, procedures for analyzing the influences of such conditions on the performance of proton exchange membrane fuel cells are expensive and time-consuming. Moreover, the commonly used models have been developed on the basis of standard ambient conditions. Thus, these models are difficult to utilize under adverse ambient conditions. This study was performed to develop suitable proton exchange membrane fuel cell models that could reflect the effects of ambient conditions on the output voltage and current of the models. The first proposed model used the advantages of electrical and thermal relationships of a complex semiempirical model of a proton exchange membrane fuel cell. A simplified proton exchange membrane fuel cell model that used passive electrical components was then developed by central composite surface design. Both proposed models were simulated using various ambient temperatures, pressures, and load resistances by considering that the applied hydrogen pressure is known. Results showed that the output voltage of proton exchange membrane fuel cell decreased when ambient temperature increased and pressure decreased. This variation was dominant when the load resistance was reduced. Computation using the simplified model was remarkably faster than that using the first model. The proposed model can be beneficial, especially for aircraft applications and unusual ambient conditions.