| Due to the rapid development of world economy and trade,people’s demand for energy continues to grow,so it is imperative to develop and utilize new clean and renewable energy sources.It is considered that hydrogen energy is the most valuable energy for R&D at present.Because it is a safe,efficient,clean and pollution-free renewable energy.PEMFC converts hydrogen energy directly into electrical energy and releases heat energy through electrochemical reaction,water is the only by-product of the reaction,and the whole reaction process does not involve the thermo-mechanical process.Therefore,it is an environmentally friendly new type of power generation device without restriction of the Cano cycle.At present,the classical 9th-order model has too high order,too many parameter variables,complex design and large amount of calculation when controlling,so it is not suitable for the research of control strategy.The analysis of PEMFC system auxiliary equipment loss and the best state of battery output is based on dynamic model only,so a control-oriented 6th-Order power generation system model including PEMFC system power output is established.Compared with the classical 9-order PEMFC system model,the output state of the 6-order model can match the output of the classical 9-order model.It only has a small error and can reflect the operation state of the PEMFC power generation system relatively accurately.This confirms the validity of the 6th order PEMFC power generation system model.The accurate estimation of oxygen saturation in PEMFC power generation system depends on the estimation accuracy of the internal unmeasurable state,so that the system can still accurately estimate oxygen saturation under external disturbances,a hybrid multi-order sliding mode observer(HMSMO)matching the system state relative degree is proposed to estimate the oxygen saturation of PEMFC power generation system.According to the difference of system state relative degree,the hybrid super-twisting(STW)sliding mode algorithm and quasi-continuous high-order(QCHO)sliding mode algorithm converge the measurement state error in a limited time,so as to ensure the internal unmeasurable variable follows the actual value of the system quickly within the error range,and realize the accurate estimation of oxygen saturation,which lays a foundation for oxygen supply regulation of the air system.The comparative experimental analysis shows that the method is insensitive to the measurement noise,can quickly track the state variables of fuel cell,has strong robustness,and then achieve accurate estimation of oxygen saturation.In order to solve the coupling problem between the parasitic loss of the air compressor and the fast supply air flow in PEMFC power generation system,and realize the net power improvement and fluctuation suppression of the system,a new method based on HMSMO and Quasi continuous high order sliding mode(QCHOSM)is proposed,Net power optimized control method(NPOCM)is used to control oxygen saturation.This method finds out the optimal operating track of oxygen saturation according to the relationship between net power and oxygen saturation and load current,Based on the oxygen saturation estimated by HMSMO,the high-order derivative of the estimated error is integrated by QCHOSM algorithm,which makes the system control quantity in a continuous state in time.This method effectively reduces the ‘buffeting’ phenomenon caused by discontinuous high-frequency control switching items to the controlled object,makes the oxygen saturation response fast,overshoot small,and accelerates the dynamic response of oxygen saturation to achieve the optimization of net power.The comparison test proves that the proposed NPOCM can accelerate the dynamic response of oxygen saturation,effectively improve the net power of the system,reduce the net power dither thickness and extend the system life. |
PDF Full Text Request