Control Strategy For DC-DC Converters Based On Switched Linear System Theory

The DC-DC converter is a basic and important device for power conversion.It plays an important role in renewable energy power generation,medical equipment charging and discharging,DC load power consumption and other fields.In essence,the DC-DC converter is a system of high nonlinearity.The traditional control strategy based on small-signal averaged model and linear system theory shows shortcomings in terms of transient performance and operation stability.To meet the higher performance requirements in practical applications,it is of great theoretical significance and engineering value to study the stability control strategy of DC-DC converters by using the new nonlinear control theory.Based on the switched linear system(SLS)theory,the DC-DC converter is characterized as a hybrid system including continuous-time and discrete-time variables.This hybrid system model is conducive to revealing the basic properties of the DC-DC converter.Furthermore,this model is useful for proposing a control strategy that has good performance and is more in line with the switching operation characteristics of the converter.However,different from general SLSs,the DC-DC converter is a class of affine SLS with uncontrollable inputs,which has not been systematically and deeply studied.Therefore,in this thesis,the basic DC-DC converters,such as the Buck-Boost converter,and the multi-level DC-DC converters are taken as research objects,and several switching control strategies are proposed for these converters by using the SLS theory.Compared with the commonly used small-signal averaged model,the SLS model accurately describes the operating characteristics of the DC-DC converter in the entire operating space,no longer limited to just considering the system characteristics around a steady-state operating point.Thus,the control research based on SLS theory provides theoretical and application references for the stability control of DC-DC converters.The detailed works in this thesis are as follows:(1)Based on the switched disturbance observer,a robust continuous-time switching control strategy for DC-DC converter is proposed.According to the SLS model characteristics of DC-DC converters,a switched disturbance observer is designed to accurately estimate the load disturbance.By using the estimated disturbance,the parameters of the switching control law are adaptively modified,so that the load disturbance is quickly and effectively suppressed and the output voltage of the converter is stabilized.It is theoretically proved that the closed-loop system consisting of the disturbance observer and adaptive switching control law ensures the Lyapunov large-signal stability of DC-DC converters.The effectiveness of the robust continuous-time switching control strategy is verified by experiments on the Buck-Boost converter.(2)A sampling-data switching control strategy for DC-DC converters is proposed,which can realize periodic time-triggered switching control according to the discretization requirements of digital control.By constructing a common quadratic Lyapunov function,the control design condition is formulated as linear matrix inequality(LMI).Compared with the existing control strategies based on bilinear matrix inequality(BMI),the complexity of LMI is lower,and there are many effective tools that can be used to solve the LMI.Experiments on the Buck-Boost converter verify the correctness and effectiveness of the sampling-data switching control strategy.(3)By using the polyhedral model description method of SLS theory,the SLS model considering the wide range of load variation is established for the DC-DC converter.On this basis,a robust sampling-data switching control strategy is proposed for DC-DC converters.Even if the load fluctuates in a wide range,the robust control strategy can always guarantee the Lyapunov exponential stability of the state variables.Moreover,all control design conditions are LMIs,which keeps the advantage of easy solution of control parameters.Experiments on the Buck-Boost converter verify the correctness and effectiveness of the robust sampling-data switching control strategy.(4)Based on the polynomial Lyapunov function and sum of squares approach of SLS theory,a switching control strategy for multilevel DC-DC converter is proposed.This strategy extends the applicability of the common quadratic Lyapunov function-based switching control strategy.It is applicable to both multilevel DC-DC converters and basic DC-DC converters,such as the Buck-Boost converter.Moreover,compared with the piecewise quadratic Lyapunov function-based switching control strategy,the proposed one has the characteristics of lower design complexity.The correctness and effectiveness of the theoretical research are fully verified by experiments on the three-level and four-level flying capacitor DC-DC converters and the Buck-Boost converter.