Proton Exchange Membrane Fuel Cell Control Research Articles

A Large-Scale Multi-Agent Deep Reinforcement Learning Method for Cooperative Output Voltage Control of PEMFCs

To increase the output voltage stability and improve the operating efficiency of proton exchange membrane fuel cells (PEMFCs), a data-driven cooperative method for controlling the PEMFC output voltage is proposed in this paper. The proposed method adapts centralized learning and decentralized implementation, which can nonlinearly and adaptively realize optimal coordinated control over the hydrogen valve and the DC/DC converter. Additionally, a champion multiagent double delay deep deterministic policy gradient (CMA-4DPG) algorithm is proposed in this method, the design of which incorporates the policies of the champion selection mechanism, cooperative exploration, imitation learning guidance and curriculum guidance to improve the robustness of the cooperative method. The method cooperates with multiple controller to prevent output voltage fluctuation. The experimental results show that by simultaneously regulating the hydrogen flow through the hydrogen valve and the duty ratio of the DC/DC converter, the proposed method achieves better robustness and can improve the tracking performance of the PEMFC output voltage.

A probabilistic analysis method based on Noisy-OR gate Bayesian network for hydrogen leakage of proton exchange membrane fuel cell

The proton exchange membrane fuel cell (PEMFC) is one of the crucial power units of the hydrogen vehicle. This work proposes a method based on the Noisy-OR gate Bayesian network to evaluate the hydrogen leakage probability of PEMFC. The generation mechanism of hydrogen leakage in fuel cells was analyzed using the fault tree theory. The conditional probability was determined using the Noisy-OR gate model in the Bayesian network, and the probability of hydrogen leakage in the fuel cell was 0.04. Through sensitivity analysis, the key factors that can cause hydrogen leaks in fuel cell were identified. The method proposed in this study can aid in PEMFC accident prevention and control strategies.

Data-driven prediction of temperature variations in an open cathode proton exchange membrane fuel cell stack using Koopman operator

In this study, a novel application of the Koopman operator for control-oriented modeling of proton exchange membrane fuel cell (PEMFC) stacks is proposed. The primary contributions of this paper are: (1) the design of Koopman-based models for a fuel cell stack, incorporating K-fold cross-validation, varying lifted dimensions, radial basis functions (RBFs), and prediction horizons; and (2) comparison of the performance of Koopman-based approach with a more traditional physics-based model. The results demonstrate the high accuracy of the Koopman-based model in predicting fuel cell stack behavior, with an error of less than 3%. The proposed approach offers several advantages, including enhanced computational efficiency, reduced computational burden, and improved interpretability. This study demonstrates the suitability of the Koopman operator for the modeling and control of PEMFCs and provides valuable insights into a novel control-oriented modeling approach that enables accurate and efficient predictions for fuel cell stacks.

Single-walled carbon nanotube interlayer modified gas diffusion layers to boost the cell performance of self-humidifying proton exchange membrane fuel cells

A tradeoff between the low humidity and the high performance remains a key challenge for the proton exchange membrane fuel cell (PEMFC). In this work, a novel self-humidifying gas diffusion layer (GDL) with a single-walled carbon nanotube (SWCNT) nonwoven layer between the gas diffusion substrate and the hydrophobic microporous layer is controllably prepared to elevate the cell performance under dry conditions. The membrane electrode assembly (MEA) with 0.25 mg cm−2 SWCNT loading exhibits a current density of 0.69 A cm−2 at 0.6 V, which is 392.8% higher than that of the counterpart without the SWCNT interlayer at the same relative humidity. Moreover, the SWCNT interlayer with rational pore structure and proper wettability dramatically improves the water retention capacity of MEA, thus enhancing the low-humidity performance of MEA. The structure design of GDL provides an effective strategy for self-humidifying PEMFC control optimization.

An LSTM and ANN Fusion Dynamic Model of a Proton Exchange Membrane Fuel Cell

Proton exchange membrane fuel cell (PEMFC) has great application prospects due to its low emission and high efficiency. An accurate model to predict the dynamic output voltage is essential for the optimal control of PEMFC for the applications on vehicles and power stations. In this paper, a novel deep learning framework with the Long short-term memory (LSTM) and ANN Fusion is proposed to develop the PEMFC dynamic model by extracting both the historical and current information. The LSTM extracts the temporal information from the past PEMFC states with its order determined with autocorrelation function (ACF) and partial autocorrelation (PACF), while the influence of the system current inputs is learnt by the ANN. Then the outputs of LSTM and ANN are concatenated with the multiple information fused to predict the PEMFC dynamic output voltage. After validated by the operating data from a lab-scale PEMFC system, the LSTM and ANN Fusion model is compared with existing models, such as SVR, ANN and LSTM methods. The comparison results show that the proposed LSTM and ANN Fusion model can provide the best prediction performance with the lowest Mean Square Error (MSE) of 1.303. The proposed LSTM and ANN Fusion model can be helpful to develop the optimal control strategy of PEMFC.

Observation and Analysis of Ejector Hysteresis Phenomena in the Hydrogen Recirculation Subsystem of PEMFCs.

The optimization control and efficiency improvement of proton exchange membrane fuel cells (PEMFCs) are being paid more attention. Ejectors have been applied in PEMFC hydrogen recirculation subsystems due to the advantages of a simple structure and no power consumption. However, the hysteresis deviation of a proportional valve ejector is found in the loading and unloading processes such that the hysteresis phenomena can cause deviations in fuel cell control process and affect the power dynamic output stability of PEMFCs. This paper analyzes the causes and effects of proportional valve hysteresis phenomena through experiments and simulations. The results show that the resultant force of proportional valve armature is different in loading and unloading processes because of the hysteresis phenomena, and the maximum flow deviation is up to 0.42 g/s. The hysteresis phenomena of flow rate further cause a deviation of 68.7-89.3 kW in PEMFC power output. Finally, a control compensation model is proposed to effectively reduce the deviation. This study provides a reference for the control and optimization of PEMFC with ejector technology.

A review of thermal management of proton exchange membrane fuel cell systems

Proton exchange membrane fuel cells (PEMFCs) have the advantages of long operation cycles, high energy efficiency, and no pollution of reaction products. Temperature is an important factor to ensure the operation of fuel cell systems. Too high temperature will cause irreversible damage to the proton exchange membrane, and too low temperature will greatly reduce the power generation efficiency of fuel cells. Therefore, the effective thermal management temperature control can ensure the stable operation of the system under steady state and dynamic variable load. It can also improve the reaction efficiency of the fuel cell system and prolong the life of the fuel cell. This paper mainly summarized the cooling mode and control strategy of PEMFCs based on thermal management system. The application of different cooling methods is further discussed. The characteristics of traditional proportional-integral-derivative (PID) control, fuzzy PID control, predictive control, adaptive control, and other common thermal management control strategies were described in detail. The research status of scholars in various countries were analyzed, respectively, and the cooling effects of different strategies are compared. Through the review and research on the temperature control of PEMFCs, it is found that the stable operation of the stack is inseparable from the coordination of reasonable cooling mode and control strategy.

Characterizing the two-phase flow effect in gas channel of proton exchange membrane fuel cell with dimensionless number

A simplified one-dimensional proton exchange membrane fuel cell (PEMFC) model is developed to characterize the two-phase flow effect with dimensionless numbers. A new Damköhler number is defined to decouple the gas and liquid flow in electrodes. The flow pattern in gas channels can be predicted by force balance analysis with Reynolds number and Weber number. The effect of two-phase flow patterns in bipolar plate gas channel on mass transfer at the gas channel-gas diffusion layer (CH-GDL) interface can be represented by Sherwood number. Then, the influence of relative humidity, cathode stoichiometric ratio, operating temperature and dimensionless width of water film/slug are studied. The results indicate that the two-phase flow occurring at high humidity and current density can be accurately predicted with dimensionless numbers. The output performance of fuel cell dropped by 10–100 mV due to the additional mass transfer resistance caused by liquid water in bipolar plate gas channels. Therefore, the present model can reflect the main two-phase transport effects with low computational cost and contribute to the efficiency and accuracy of PEMFC design and control.

A critical survey of proton exchange membrane fuel cell system control: Summaries, advances, and perspectives

Proton exchange membrane fuel cell (PEMFC) is a promising future power source, which uses hydrogen energy to generate electricity with the byproduct of water. In general, PEMFC includes several strongly coupled subsystems with high nonlinearity and complex dynamic processes. Therefore, proper control strategies are crucial for a reliable and effective PEMFC operation. This paper aims to carry out a comprehensive and systematic overview of state-of-the-art PEMFC control strategies. Based on a thorough investigation of 180 literatures, these control strategies are classified into nine main categories, including proportional integral derivative (PID) control, adaptive control (APC), fuzzy logic control (FLC), robust control, observer-based control, model predictive control (MPC), fault tolerant control (FTC), optimal control and artificial intelligence control. Furthermore, a comprehensive evaluation and detailed summary of their control deigns, objectives, performance, applications, advantages/disadvantages, complexity, robustness and accuracy are conducted thoroughly. Finally, five valuable and insightful perspectives/recommendations are proposed for future research. • Summarize nine groups of PEMFC system control strategies. • Control complexity, robustness, and accuracy are summarized and evaluated. • Present the advantages/disadvantages of all methods. • Five invaluable and insightful perspectives/recommendations are proposed.

Progress of proton exchange membrane fuel cell

The rapid development of the economy of China leads to the significant growth of energy production and consumption demand. At present, the total amount of greenhouse gas emissions has ranked second in the world. To meet the challenges caused by global climate change and environmental pollution as well as realize the sustainable development of our country’s economy, carbon neutrality should be achieved. As a new type of low-temperature fuel cell, proton exchange membrane fuel cell (PEMFC) attracts a lot of attention due to its safe and environmentally friendly characteristics. However, the operating efficiency of the current commercial PEMFC is rather low (about 50%), which means that about half of the energy will be released as thermal energy. The stable and reliable heat dissipation of the fuel cell has a decisive impact on its performance and life, i.e., thermal management issues are particularly important for the efficiency and stability of the PEMFC stack. This article reviews the thermal management of PEMFC and analyzes the specific applicable conditions of cooling methods as well as the latest developments in thermal management methods. Specifically, it focuses on the thermal characteristics of PEMFC and its effects on performance and life. Besides, the PEMFC thermal management schemes and control strategies under different cooling methods are also summarized.

Decentralized event-triggered adaptive control for interconnected nonlinear dynamics of constrained air supply and thermal management systems of PEMFCs

This paper investigates a decentralized event-triggered adaptive control problem for uncertain interconnected air supply and thermal management nonlinear dynamics of proton exchange membrane fuel cells (PEMFCs). In addition, the constraint problems of the oxygen excess ratio (OER) and stack temperature (ST) are addressed for preventing oxygen starvation, parasitic loss, membrane dehydration, and electrode flooding of PEMFCs. First, the practical nonlinear interaction problem between the air pressure of the air supply subsystem and the ST of the thermal management subsystem is formulated. Then, a decentralized adaptive control design strategy is established to guarantee the constraint satisfaction of each subsystem and the asynchronously event-triggered regulation of the OER and ST. In contrast to the previous studies on the control of PEMFCs, this study first addresses the nonlinear interaction problem of the air supply and thermal management systems and their decentralized control design in the control field of PEMFCs. Based on the Lyapunov stability theorem, the stability of the resulting overall control scheme is analyzed.

Air Supply System Control of PEM Fuel Cell for Electric Vehicle Application Based on Multi-layer Prediction Strategy

Proton exchange membrane (PEM) fuel cell engine has many advantages, including high energy density, high efficiency, low operation temperature, and zero emissions, which is a promising application to electric vehicles. In this paper, a multi-layer prediction control strategy for air supply system of a full-power PEM fuel cell electric vehicle is proposed, and a mathematical model of PEMF fuel cell engine air supply system is established in MATLAB/Simulink environment. The control scheme of the proposed multi-layer prediction strategy is that the top-layer model prediction is used for predicting the driving condition (speed of the vehicle) to obtain the desired air mass flow rate, and the bottom-layer air flow model prediction control (MPC) can adopt the top-layer airflow demand to regulate the oxygen excess ratio of PEM fuel cell engine. The proposed control strategy can meet the needs of the fuel cell stack reaction of oxygen as well as prevent air starvation that might occur in PEM fuel cell electric vehicle during driving conditions variation.

Modeling, control and simulation of a photovoltaic /hydrogen/ supercapacitor hybrid power generation system for grid-connected applications

Hybrid renewable energy systems (HRES) should be designed appropriately with an adequate combination of different renewable sources and various energy storage methods to overcome the problem of intermittency of renewable energy resources. Focusing on the inevitable impact on the grid caused by strong randomicity and apparent intermittency of photovoltaic (PV) generation system, modeling and control strategy of pure green and grid-friendly hybrid power generation system based on hydrogen energy storage and supercapacitor (SC) is proposed in this paper. Aiming at smoothing grid-connected power fluctuations of PV and meeting load demand, the alkaline electrolyzer (AE) and proton exchange membrane fuel cell (PEMFC) and SC are connected to DC bus of photovoltaic grid-connected generation system. Through coordinated control and power management of PV, AE, PEMFC and SC, hybrid power generation system friendliness and active grid-connection are realized. The validity and correctness of modeling and control strategies referred in this paper are verified through simulation results based on PSCAD/EMTDC software platform.

An Enhanced Fuel Cell Dynamic Model With Electrochemical Phenomena Parameterization as Test Bed for Control System Analysis

In this work, a model of a proton exchange membrane fuel cell (PEMFC) is presented. A dynamic performance characterization is performed to assess the cell transient response to input variables. The model used in the simulation considers three different phenomena: mass transfer, thermodynamics, and electrochemistry. The main sources of voltage loss are presented: activation, electrical resistance, and concentration. The model is constructed to avoid the use of fitted parameters, reducing the experimentation required for its validation. Hence, the electrochemical model is parameterized by physical variables, including material properties and geometrical characteristics. The model is demonstrated as a test-bed for PEMFC control system design and evaluation. Results demonstrate that the steady-state and dynamic behavior of the system are represented accurately. A case study is included to show the functionality of the model. In the case study, the effect of the purge valves at the fuel cell discharges is analyzed under different scenarios. Regular purges of the cathode and the anode are shown to achieve a good performance in the system avoiding reactant starvation in the cell. A closed-loop dynamic response is included as an example of the model capabilities for the design of fuel cell control strategies. Two variables were selected to be controlled: voltage and pressure difference across the membrane. A multivariate control strategy was tested and its dynamic response was analyzed. It was found that there was a strong interaction between the control loops, making the control of the system a challenge.

Oxygen Excess Ratio Control of PEM Fuel Cell Based on Self-adaptive Fuzzy PID

Oxygen excess ratio is closely related to the output efficiency and service life of the proton exchange membrane (PEM) fuel cell system. To simulate dynamic behaviors of air flow, a fourth-order model of PEM fuel cell system is established. Then, a self-adaptive fuzzy PID (SFPID) controller is proposed to regulate the oxygen excess ratio on-line through its adaptive characteristic. Under different current disturbances, a comparison of proposed SFPID with several control topologies such as feedforward, PID, and PID plus feedforward (PID-FF) is carried out to validate the advantages of the proposed controller. Moreover, the proposed SFPID controller is applied to track the variable oxygen excess ratio. The simulation results display that the set-point value of oxygen excess ratio can be tracked rapidly and accurately, and verify that the proposed SFPID controller has good dynamic response performance.

Control structure design and robust model predictive control for controlling a proton exchange membrane fuel cell

In this work, cell performance analysis, concept of control structure design, and controller design for a proton exchange membrane fuel cell (PEMFC) were implemented. Steady-state analysis was performed to determine suitable operating conditions. The effects of the input parameters on cell voltage and cell temperature were analyzed to investigate the dynamic behavior of a PEMFC that is important for control design. To obtain an efficient control system, the control structure design of the PEMFC was also considered to find suitable controlled and manipulated variables. Moreover, the selection of input-output pairings by considering the relative gain array (RGA) as a controllability index was applied to the control system design. Finally, model predictive control and offline robust model predictive control (MPC) based on a linear time-varying model were proposed for PEMFC control. The results showed that the steady-state operating points were selected at the current density = 0.51 A/cm2, cell voltage = 0.59 V, power density = 0.30 W/cm2, and cell temperature = 332 K. The cell voltage and cell temperature depend on the inlet molar flow rates, temperature of hydrogen and air, and operating current density. According to the RGA, the inlet molar flow rates of air and hydrogen are manipulated variables that regulate the cell temperature and partial pressure of hydrogen, respectively. Furthermore, controller design using MPC and robust MPC as a controller can demonstrate good results. Robust MPC can guarantee the stability of the PEMFC.

Combined predictions for prognostics and predictive control of transportation PEMFC

Abstract: To help a transition of prognostics approaches toward industries, it is necessary to show that they can be adapted in every situation. Nowadays, a lot of prognostics applications focus on energy sources, among them Proton Exchange Membrane Fuel Cells (PEMFC) can be cited. Due to their wide range of applications, different prognostics adaptations should be considered. Issues coming with PEMFC used for transportation are considered in this paper. Different time scales are involved, requiring a modification of the existing approaches. This paper proposes a solution to perform short-term and long-term predictions on a PEMFC stack used in a transportation application based on particle filters. After proposing different data reductions, the adapted particle filters configuration for this use case is determined. Accurate State of Health (SoH) estimations and predictions, with high coefficient of determination, are obtained. Behavior predictions are also performed and show promising results.

The development of a hybrid PEMFC power system

This paper describes the development of a stationary hybrid power system consisting of a 3 kW proton exchange membrane fuel cell (PEMFC) module, a battery set, and solar cells. We design PEMFC control and power managements for the hybrid system, and construct a simulation model based on experimental data. The simulation model is then used to analyze performance and efficiency of the system, employing different power management strategies. The study comprises three parts: PEMFC control, power management, and the simulation model. First, we apply robust control to regulate the PEMFC temperature and current, and achieve root mean square errors of 0.06 °C and 0.07 A, respectively. Second, we integrate the hybrid power system and design power management strategies for sustainable operation. Third, we develop a Matlab™ SimPowerSystem model based on experimental data. The simulation results are shown to match the experiments with a deviation of 0.21%. Last, we apply the simulation model to discuss the impacts of different power management strategies on system efficiency. Based on the results, the designed controllers and power management strategies are deemed effective at improving system performance and efficiency. In addition, the developed simulation model can be applied in the future to develop customized hybrid power system.

Thermal management system modeling of a water-cooled proton exchange membrane fuel cell

In water-cooled proton exchange membrane fuel cell (PEMFC) system, a thermal management system which involves coordinating thermal management components such as the coolant circulation pump and radiator fan plays the important role for keeping the output performance of PEMFC in safe and steady condition. In order to avoid coolant temperature fluctuations and prevent thermal runaway, a control-oriented water-cooled system model based on electrochemical reactions and thermodynamic for the thermal management control is established and an experimental system is also developed for model validation. The comparisons between the model results and the experimental data under three different operating conditions (increasing the coolant temperature difference between outlet and inlet, reducing the coolant inlet temperatures, increasing the output current) demonstrate that the proposed model can efficiently simulate the dynamic behaviors with a high accuracy for the changing of the output voltage, the coolant temperature, the coolant flow rate and the controlling voltage of radiator. Therefore, the proposed model will be used as guidance for the design and optimization of the water-cooled PEMFC control system.

Robust control of nonlinear PEMFC against uncertainty using fractional complex order control

This work proposes a fractional complex order controller (FCOC) design strategy to cope with uncertainty in a proton exchange membrane fuel cell (PEMFC) model. The fuel cell dynamic behavior is inherently nonlinear and time varying. Accordingly, a locally linearization technique is used to achieve a linear interpretation in form of transfer function instead of nonlinear dynamics. When the current load is suddenly changed, the voltage and consequently the operating point are dramatically varying. Therefore, the resultant linearized model of the PEMFC changes. The discrepancy between those deviated models from the nominal plant will be regarded as system uncertainties, which must be cured by robust controller. In PEMFC dynamic, the ratio of the oxygen with respect to the air supply, i.e. the oxygen excess ratio (\( \lambda _{{\mathrm{o}_{2} }}\)), is required to be adjusted. A sudden load variation causes huge variations in \( \lambda _{{\mathrm{o}_{2} }}\). Main purpose of this manuscript is to investigate the capability of the FCOC to regulate \( \lambda _{{\mathrm{o}_{2} }}\) in different operating conditions. The designed controller will be gained to satisfy multi-constraint problem. The performance of the controllers is verified in the presence of uncertainty by means of the frequency criteria, i.e. the phase and the gain margins, as well as the time indices. The quality of the controller will be investigated on the original nonlinear plant. The stability and performance of the proposed controller with respect to other conventional controllers, e.g. \(PI\), fractional order \(PI\)(FO-PI) and \(\hbox {H}_\infty \) will be also investigated.