| As key interfaces of wind energy conversion and power conversion in the renewable energy distributed generation(DG)system,variable-speed wind turbine and inverter are two core devices in the DG system.However,under the complicated operating environment,due to the existence of practical engineering factors such as nonlinearity and parameter uncertainty of variable-speed wind turbines,variation of inverter filter parameters,nonlinear local loads,unbalance local loads and local loads varying with system working conditions,the linear control method represented by PI cannot meet the high-performance control requirements for these two core devices in the modern DG systems.Advanced control methods are effective means to solve the problem existing in linear control method.So,this thesis takes the variable-speed wind turbine,the single-phase and three-phase inverters(named grid-side inverters for grid-connected operation and load-side inverters for off-grid operation)as research objects.Meanwhile,according to their different control modes in the DG system application,this thesis conducts in-depth research on related control methods based on representative advanced control theories such as sliding mode control,predictive control and adaptive control.The main research contents are as follows:(1)Aiming at the problem that the sliding mode chattering can cause high-frequency oscillations in the control torque of the variable speed wind turbine,the traditional constant velocity approaching law is analyzed,the contradiction between its approaching speed and sliding mode chattering level is pointed out.An improved constant velocity approaching law is proposed and it reduces the sliding mode chattering level while accelerating the approaching speed.Based on this improved constant velocity approaching law,the process of the sliding mode controller for the variable speed wind turbine is designed.Based on the Lyapunov stability theory,an aerodynamic torque observer is designed.The value range of switching gain is reduced by using feedforward compensation of the aerodynamic torque,and the sliding mode chattering can be further suppressed.Based on the above,a maximum power point tracking sliding mode approach law control system for variable speed wind turbine is constructed with the tip speed ratio method,and the performance of system control is improved.(2)Aiming at the problem that the model predictive current control has poor robustness to inductance parameters,a robust constant frequency model predictive current control method for grid-side inverters is proposed.The optimal time series constant frequency model predictive control method is adopted as the framework,which can ensure a fixed switching frequency.In order to improve the robustness of the current prediction model,a robust current prediction model is obtained by adding robust terms,prediction error feedback terms and inductance feedforward compensation terms to the current prediction model,which can accelerate convergence speed of the prediction model,and can realize strong robustness to inductance parameters,which reduces sensitivity to inductance parameters for the model prediction current control.Based on the relationship between inductive voltage and current,an inductance estimator with clear physical meaning,simple structure and fast response is designed.(3)Due to changing inductance parameters has great impact on steady-state and dynamic performance of deadbeat direct power control,a robust deadbeat direct power control method for grid-side inverters is proposed.A simplified power model of the grid-side inverter is established.Based on this model,a power disturbance observer is designed.When inductance parameters change,the steady-state performance of the system is guaranteed by power disturbance feedforward compensation.The inductance estimator is designed based on the power disturbance model,and adjustment of inductance parameters is realized online,which avoids the influence of non-precision inductance parameters on the transient performance of the system.Above methods can ensure best control performance for active and reactive power when the inductance parameter changes.The controller is designed by the simplified power model and the power prediction is carried out by the power disturbance observer,it avoids the increase of the algorithm’s computational cost.(4)Taking the load-side inverter in the single-phase DG system as research object,three output voltage control methods are proposed to reduce the influence of load changing,LC the uncertainty of filter parameters and nonlinear load on voltage waveform quality,details as follows:(a)In view of the problem that the local load varies with the working conditions of the system,this thesis has studied an output voltage control method based on load current sliding mode observer for the load-side inverter.Based on the cascade control theory,a sliding mode controller for voltage outer loop and a proportional controller for current inner loop are designed.Based on the principle of extended observer,a load current sliding mode observer is proposed which improves the adaptive ability of the system to local load and avoids the influence of sliding mode chattering on voltage waveform quality.The proposed load current sliding mode observer has low-pass filtering properties and can be directly applied to engineering practice.(b)The proposed output voltage control method based on the load current sliding mode observer for the load-side inverter does not have good robustness to LC filter parameters.For this reason,in view of the uncertainty of LC filter parameters,an output voltage control method based on backstepping sliding mode for the load-side inverter is proposed.In the last step of the traditional backstepping design,the robustness to LC filter parameters and the anti-disturbance ability to local load are improved by adding the sliding mode robust term,and the single closed loop voltage control is also realized.(c)The proposed output voltage control method based on backstepping sliding mode for the load-side inverter requires prior knowledge of local load parameters and LC filter parameters.For this reason,this thesis has proposed an output voltage adaptive complementary sliding mode control method for the load-side inverter.Firstly,This method designed the output voltage complementary sliding mode controller,and then designed a LC filter parameter adaptive law and an inductor current estimator respectively.A single closed-loop output voltage adaptive control without prior knowledge of local load parameters and LC filter parameters is realized.(5)For unbalance and nonlinearity of the three-phase local load and LC filter parameter uncertainty,an adaptive output voltage control method for the load-side inverter in the three-phase DG system is proposed.Based on the dq model of load side inverter and taking output voltage and its derivative as system state variables,a dq model for single closed loop voltage control is obtained.The system disturbance is divided into two parts:steady-state disturbance and dynamic disturbance.The PID controller,steady state disturbance adaptive law and robust control term are designed respectively.The PID controller forces the voltage tracking error to zero,the steady-state disturbance adaptive law compensates the steady state disturbance online,and the robust control term suppresses the dynamic disturbance.The proposed control method does not require phase sequence decomposition.High performance output voltage control for any type of local load and strong robustness to LC filter parameters are realized only by single closed loop voltage control. |