A novel robust reset controller implemented through field programmable gate array for oxygen ratio regulation of proton exchange membrane fuel cell
Introduction
The problems of energy supply and global warming are taking on new dimensions every day, which have led to further efforts to operationalize and optimize renewable energy sources and energy storage [1], [2], [3]. One of these sources is fuel cell which is a suitable solution for producing clean and efficient energy by converting chemical energy into electrical energy [4]. There are currently several types of fuel cells, which one of the most common is the Proton Exchange Membrane (PEM) fuel cell and it generates heat, water and electrical energy through hydrogen and oxygen reactions [5]. High power density, high efficiency, low operating temperature, low noise and less pollution are some of PEM fuel cell advantages.
In general, four main subsystems impact on PEM fuel cell, including air supply subsystem, hydrogen feeding subsystem, humidity management subsystem and temperature management subsystem and for the same reasons, the dynamics of PEM fuel cell is non-linear in nature. Hence, for effective and optimal operation besides increasing lifespan of PEM fuel cell, it is necessary to take the precise and correct control measures for each of the subsystems [6]. According to previous studies, the efficiency and output power of PEM fuel cell are directly affected by the air (oxygen) supply system and therefore one of the fields of study is the controller design in relation to the PEM air supply subsystem. It should be noted that any deficiency in oxygen supply can lead to voltage and output power degradation, stack flooding, membrane failure and also can reduce the life of the fuel cell [7].
To date, different control methods have been used for the PEM fuel cell air supply system using different models. The ninth-order model [8], the simplified transient model [9], the 4th order model [10] and the 3rd order model [11] are among the models used to design the controller. The most important control methods designed can be mentioned as follows: Linear Quadratic Regulator (LQR) [12], Linear Quadratic Gaussian (LQG) [13], feedback linirization [14], Model Predictive Control (MPC) [15], sliding mode [16] and different types of Proportional Integral Derivative (PID) [17], [18] are among the methods used to keep the oxygen rate at an optimal value. Also, oxygen starvation has been prevented through stoichiometric control of oxygen by an adaptive method in [19], and the neural network method has been used to control the current demand using the parametric cerebellar model [20], which has a functional advantage over the conventional PI control method.
Despite the appropriate theoretical results, most of the control methods used suffer from high complexity and do not have the ability to be implemented in practice. Also, measurement error and signal fluctuations in the results obtained from linear models make serious challenges for the efficiency of control methods. Then, it is difficult to ensure the stability of the closed loop system. Therefore, the main innovations of this study are: 1. Considering nonlinear dynamics in controller design to more accurately describe PEM fuel cell behavior; 2. Providing a control approach with a simple and efficient structure to achieve the goal of regulating the oxygen rate and implementing on the fuel cell system in practice; 3. Fast and smooth tracking and control of the optimal oxygen rate for the PEM fuel cell despite the most severe possible changes in the stack current caused by various parameters without knowledge about the amount of current variations; 4. Implementing the proposed controller structure through FPGA, which enables operational and practical implementation of the controller.
To this end, this paper presents an innovative approach using the reset method with the ability for implementing through Field Programmable Gate Array (FPGA) to control the oxygen rate of Proton Exchange Membrane Fuel Cell (PEMFC). The reset method that has been considered recently [21], [22] is able to overcome the limitations of linear controllers despite its simple structure and can guarantee zero error in tracking the desired amount of oxygen rate. Using the reset mechanism that is triggered when the tracking error is zero, this control method has been planned to achieve the control goal in this paper. In addition, FPGA is used as a suitable platform for embedded control systems [23], [24]. This hardware-based platform provides advantages such as high execution speed, flexible design and high reliability [25]. It also allows the designer to explore and verify the control algorithm and, optimize its performance in less time and cost without real and practical implementation. Considering the advantages of FPGA in different areas of performance, reliability, long term maintenance, cost, and time to market [26], this paper presents a new robust control method using the reset method implemented through FPGA to regulate PEM fuel cell oxygen rate.
Accordingly, this article has been compiled as follows; in the second part, the model of PEM fuel cell system and the equations governing air supply subsystem are given and then the control goal is stated. In the third part, the proposed FPGA reset control structure is described and the implementation of its various parts is fully described. Section 4 shows the results obtained from the proposed method and its comparison with the PID fuzzy hybrid method. Finally, in the fifth part, the conclusion of the article is given.
Section snippets
Nonlinear model of PEM fuel cell air supply system
This section describes the model used for controller design. Recent advances have led to the development of a dynamic control-oriented model to describe PEM fuel cell behavior. This type of model is necessary for control structure development of PEM fuel cell systems in which there are unpredictable and variable variations in power demand. In dynamic modeling, conservation equations for momentum, mass, species, charge, and energy are used to describe heat transfer and temperature distribution
The proposed FPGA reset controller
This section describes the structure of the FPGA reset controller. Fig. 2 shows the overall structure of the controller. According to Eq. (5) and other relationships in Section 2, the measured output oxygen rate of the PEM fuel cell () is compared with the desired oxygen rate (), and the error obtained is used as the input of the FPGA reset controller. Then, the control signal obtained from the proposed approach is applied to the PEM air supply subsystem. Now the different parts of the
Simulation results
To demonstrate the efficiency of the proposed FPGA reset method, the results obtained from its implementation are compared with the results obtained from the hybrid fuzzy PID method [18]. For this purpose, the four commonly criteria are considered including Integral Squared Error (ISE), Integral Absolute Error (IAE), Integral Time Absolute Error (ITAE), and Integral Time Square Error (ITSE). In PEMFC, stack current is affected by several parameters and factors and for evaluating the robustness
Conclusion
Because of the increasing importance of renewable energy, especially fuel cells in the power supply chain and the reduction of environmental pollution, the issue of proton exchange membrane fuel cell control was considered in this paper. First, a nonlinear model was proposed for the air supply subsystem, and due to the importance of controlling the oxygen rate to provide maximum output power and guarantee its longlife, as well as to prevent membrane degradation, a new robust control method was
Declaration of Competing Interest
None
WU Linli is with Academy of Information Technology, Luoyang Normal University, Luoyang Henan Province, China. She received the B.Sc degree from Xinjiang University, China, and the M.Sc degree from Henan University of Science and Technology, China. Her current research interests include artificial intelligence, machine learning, and neural networks.
References (30)
- et al.
Ultrathin MoS2 nanosheets anchored on carbon nanofibers as free-standing flexible anode with stable lithium storage performance
J Alloys Compd
(2022) - et al.
PEM fuel cell system control: A review. Renewable Energy
(2017) - et al.
Fuel cell flooding diagnosis based on time-constant spectrum analysis
Int J Hydrogen Energy
(2016) - et al.
A new dynamic model for predicting transient phenomena in a PEM fuel cell system
Renewable Energy
(2005) - et al.
Passivity and robust PI control of the air supply system of a PEM fuel cell model
Automatica
(2011) - et al.
Design and experimental validation of a constrained MPC for the air feed of a fuel cell
Control Eng Pract
(2009) - et al.
Real-time implementation of a neural model-based self-tuning PID strategy for oxygen stoichiometry control in PEM fuel cell
Int J Hydrogen Energy
(2014) - et al.
Novel hybrid fuzzy-PID control scheme for air supply in PEM fuel-cell-based systems
Int J Hydrogen Energy
(2017) - et al.
Control and real-time optimization of an automotive hybrid fuel cell power system
Control Eng Pract
(2009) - et al.
Dynamic neural network controller model of PEM fuel cell system
Int J Hydrogen Energy
(2009)
Loop-shaping for reset control systems: A higher-order sinusoidal-input describing functions approach
Control Eng Pract
Reset observer-based fault tolerant control for a class of fuzzy nonlinear time-delay systems
J Process Control
Fundamental properties of reset control systems
Automatica
A novel potassium-ion-based dual-ion battery
Adv Mater
Integration of hydrogen storage system and wind generation in power systems under demand response program: A novel p-robust stochastic programming
Int J Hydrogen Energy
Cited by (2)
Dynamic performance improvement of proton exchange membrane fuel cell system by robust loop shaping and artificial intelligence optimized fractional order PI controllers
2023, Energy Sources, Part A: Recovery, Utilization and Environmental Effects
WU Linli is with Academy of Information Technology, Luoyang Normal University, Luoyang Henan Province, China. She received the B.Sc degree from Xinjiang University, China, and the M.Sc degree from Henan University of Science and Technology, China. Her current research interests include artificial intelligence, machine learning, and neural networks.
SHEN Zhangyi is an Assistant Professor of Computer Science at School of Computer and Information, Anhui Normal University, China. He completed his Ph.D. at Shanghai University, and experienced post-doctoral at Harbin Institute of Technology, Shenzhen. His research interests are artificial intelligence, big data analysis, and decision supporting.
Hassan Mousavi is a Ph.D student in Control and System Engineering and his research interests include Nonlinear Control, MPC, Robust control and Renewable Energy.
Hai Gu was born in Zhenjiang City, Jiangsu Province. P.R. China, in 1982. He received the master's degree from Nanjing University of Aeronautics and Astronautics, P.R. China. He works in Nantong Institute of Technology.His research interest include intelligent manufacturing and additive manufacturing.
This paper was submitted for regular issues of CAEE, but should be included in special section VSI-fpga3. Reviews were processed by Area Editor Dr. E. Cabal-Yepez and recommended for publication.