Model-Free Robust Control And Energy Management Strategy For Fuel Cell Power Systems With Nonlinearities

Solid oxide fuel cells(SOFCs)and proton exchange membrane fuel cells(PEMFCs)have received a large amount and increasing attention in the research area of fuel cells systems and are widely used in different applications because of their high energy conversion efficiency and environmental compatibility.The SOFC and PEMFC systems are highly nonlinear systems with operating constraints,external disturbances,unmeasurable variables,parameter and model uncertainties.In order to obtain a high-quality power supply from these systems with optimal dynamic performance compromising between safety and efficiency,advanced control techniques are required.Besides,the transient performance is a key characteristic of the PEMFC/battery hybrid power system.Particularly,in the vehicular application,a PEMFC stack is typically coupled with a battery through DC/DC converters to remedy the drawback of the slow dynamic response of PEMFC and to deliver reliable power.Due to the difference in dynamics prosperities between the PEMFC,the battery and DC/DC converters,the PEMFC/battery hybrid power system is structurally complex with strong nonlinearities and uncertainties.In addition,these uncertainties and nonlinearities may lead to performance degradation,instability of the closed-loop system,and reduce the power source’s lifespan.To deal with this problem,a high efficient energy management strategy is required.Thus,the priority of this dissertation is to develop adaptive intelligent model-free controllers for SOFC and PEMFC systems and an optimal energy management strategy for the PEMFC/battery hybrid power system to achieve the required performance.The main contributions of this dissertation are given as follows:Firstly,for the nonlinear SOFC system with magnitude constraint of control input,parameter uncertainties,and external disturbances,a new output feedback based-intelligent proportional integral-adaptive sliding mode control(i PI-ASMC)is proposed.The presented i PI-ASMC scheme is designed based on an extended state observer(ESO)and anti-windup compensator to regulate the output voltage of the SOFC system.The ESO observer is used to estimate the unknown system dynamics,while the anti-windup compensator is employed to deal with the actuator saturation,which is caused by magnitude constraint.The Lyapunov theory is used to analyze the stability of i PI-ASMC via the closed-loop system.Moreover,the effectiveness of the proposed i PI-ASMC controller is assessed with the PID,i PI,and fuzzy PID controller,and the simulation results are presented.Secondly,for the nonlinear SOFC system with both magnitude and rate constraints of control input and measurement noise,a novel output feedback based-enhanced model-free discrete-time adaptive terminal sliding-mode control(EMF-ATSMC)is developed.The referred EMF-ATSMC controller is composed of three parts: the pseudo-partial-derivative(PPD)estimator,the discrete-time adaptive terminal sliding-mode control,and the anti-windup compensator.The PPD estimator is designed based on an enhanced compact form dynamic linearization(CFDL)data-driven modeling for the SOFC system,considering the load perturbation.Then,the discrete-time adaptive terminal sliding-mode control via an anti-windup compensator is developed to improve the control performance and guarantee the controlled system stability,wherein the designed anti-windup compensator is employed to eliminate the magnitude and rate saturations of control input.Furthermore,by means of the Lyapunov method,the stability of the proposed EMF-ATSMC via the closed-loop system is theoretically verified.The obtained simulation results are presented to demonstrate the dynamic performance of the proposed EMF-ATSMC controller.Thirdly,for the nonlinear PEMFC system with unmeasurable state variables,external disturbances and parameter uncertainties,a novel nonlinear disturbance observer based-modelfree adaptive interval type-2 fuzzy sliding mode control(MF-AIT2 FSMC)is proposed.The main objective of the designed MF-AIT2 FSMC approach is to regulate the oxygen excess ratio accurately,aiming to avoid oxygen starvation and obtain maximum output net power.The presented MF-AIT2 FSMC technique comprises three sub-components: the first sub-component,nonlinear disturbance observers(NDBO),is proposed to estimate unmeasurable state variables.The second sub-component,NDBO based-intelligent proportional integral controller,is used to estimate the unknown system dynamics via the knowledge of control input and output signals.The third sub-component,an adaptive interval type-2 fuzzy nonsingular fast terminal sliding mode control,is added to the NDBO-i PI controller to compensate for the NDBO estimation error,enhance the control performance and ensure the global controlled-system stability.Furthermore,the stability of the proposed MF-AIT2 FSMC via a closed-loop system is verified using the Lyapunov approach.Finally,the obtained simulation results are presented to demonstrate the robustness and efficiency of the proposed MF-AIT2 FSMC controller.Finally,for the PEMFC/battery hybrid power system with equivalent fuel consumption,PEMFC degradation and two typical driving cycles,a new energy management strategy based on the bat-optimized fuzzy controller with fractional-order adaptive super-twisting sliding mode control(FO-ASTSMC)is developed to improve the fuel economy and prolong the lifetime of the PEMFC system.The proposed bat-fuzzy with FO-ASTSMC scheme is employed to focus on optimal power allocation and stabilize the DC bus voltage.The bat optimization algorithm is employed to obtain optimal parameters of fuzzy membership functions(MFs)while minimizing the target cost function,considering the equivalent hydrogen consumption with PEMFC degradation.Then,FO-ASTSMC control loops are proposed to control the fuel cell and battery currents to follow their given reference values.The stability of the proposed FO-ASTSMC via a closed-loop system is theoretically verified using the Lyapunov theory.The PEMFC/battery hybrid power system simulation model is established on MATLAB/Simulink,and highway fuel economy test(HWFET)and federal test procedure(FTP)driving cycles are used for the investigation.