Study On Online Parameter Estimation Technique Of Buck Converter Main Circuit

With the increasingly widespread use and emerging critical applications of power electronics,the losses caused by power electronic converter failures have become unacceptable.Reliability of power electronic converters has been a key issue in the field of power electronics.Condition monitoring is an important mean to improve the operation reliability and reduce maintenance and operation costs in power electronic systems,while online estimation of main circuit parameters can provide effective information for condition monitoring.Novel online estimation methods of power electronic converter main circuit parameters are studied in this thesis.Capacitors are one kind of most fragile components in converters.A non-invasive online estimation method for Buck converter output capacitor parameters is proposed in this thesis.By analyzing the steady-state waveforms of non-ideal Buck converters working in the continuous conduction mode(CCM)and the discontinuous conduction mode(DCM),the relationships between capacitor parameters and output voltage ripple are obtained.Based on these relationships,various capacitance and equivalent series resistance(ESR)estimation approaches for different estimation conditions and sampling constraints are proposed.To build a complete online estimation system,the steady state criterion,work mode criterions,and reference design flow are given.To verify the proposed method,a prototype has been made and tested.The experimental results validate the effectiveness of the method and the reasonableness of the reference design flow.Estimating parameters of several components together can provide more extensive information of converter health condition.A low sampling rate and non-invasive online parameter estimation method for converters using observer is proposed in this thesis.Firstly,a parameter observer algorithm for discrete-time systems is proposed based on based on Lyapunov stability theory of discrete-time switched systems.Then,the discrete-time modeling approach for converters is introduced taking the CCM Buck converter as an example.On this basis,a parameter observer which is able to simultaneously estimate the capacitance,capacitor's ESR,inductance,inductor's ESR,and equivalent load resistance is proposed.Finally,both simulation and experimental results are presented to validate the effectiveness of the proposed estimation method.The proposed method has a low sampling rate requirement and can estimate parameters without affecting converter operation.Furthermore,it is easy to be generalized to different converters.