Research On Fractional Pid Control And Optimization Simulation Of Photovoltaic Grid Connected MMC

Modular Multilevel Converter(MMC) can connect multiple sub-modules in series to adapt to different voltage levels without being limited by power device capacity.However,due to problems such as the circulating current of the MMC and the voltage equalization of the sub-module capacitance,the cost and power consumption increase,which makes the control more complicated.The tracking accuracy of the traditional PID controller is not high,and the adjustable range of the fractional order PI~λD~μcontroller is wider.Theoretically,a better control effect should be obtained.This thesis uses the fractional order PI~λD~μcontrol strategy to design the MMC grid-connected system.The main work is as follows:Firstly,MMC is modeled,its working principle and mathematical model are analyzed,and the relevant parameters in the internal circuit of MMC are calculated.Based on the fractional order PI~λD~μcontroller,the circulation suppression strategy and the sub-module capacitor voltage control strategy are designed.The characteristics of photovoltaic cells are analyzed.A power feed-forward double closed-loop control strategy based on fractional order PI~λD~μcontroller is designed.The fractional order PI~λD~μcontroller is used in the inner loop to accurately track the fundamental frequency current,and the power feed-forward control is introduced in the outer loop to improve the voltage tracking speed.Secondly,the theoretical basis of fractional calculus is introduced,and several approximation methods are fully compared.The improved Oustaloup approximation method with better approximation effect is selected to design the controller.The approximate effects of the Oustaloup filter in different orders and frequency bands are fully compared,and the optimal parameter selection principle of the Oustaloup filter when designing a fractional-order controller is given.Then the improved Oustaloup approximation method is used to design the fractional PI~λD~μcontroller,and the influence ofλandμvalues in the fractional PI~λD~μcontroller on the system and its regularity are analyzed.Then,the system model is built in Simulink,and the circulation suppression strategy based on integer order PID controller and fractional order PI~λD~μcontroller,as well as the control strategy of sub module capacitance voltage are simulated and compared.It is proved that fractional order PI~λD~μcontroller can better suppress the harmonic in circulation,improve the distortion rate of output current and voltage,and have better control effect.The simulation of PV grid connected system is carried out,and the control system based on the traditional PID controller was compared with the power feedforward dual closed-loop control system based on the fractional order PI~λD~μcontroller.The tracking speed of the voltage change on the DC side of the inverter is verified.The results verify that the tracking speed of the power feed-forward double closed-loop control system based on fractional order PI~λD~μcontroller is faster.Simulation analysis proves that the control system has strong robustness.Finally,fuzzy algorithm and improved particle swarm optimization are used to optimize the parameters of fractional order PI~λD~μcontroller.The simulation results show that both algorithms optimize the parameters of fractional order PI~λD~μcontroller.Relatively speaking,the optimization effect of fuzzy algorithm is better.