Control Strategy Of MMC Photovoltaic Grid-connected System Under Grid-side Asymmetric Fault

In recent years,fossil energy has been increasingly consumed,and renewable energy including photovoltaic power generation has become a research hotspot.Modular multi-level converter is a new type of converter suitable for high-voltage and high-power working environments.Using it as a photovoltaic inverter has the advantages of easy modular control and reduction of harmonic distortion rate.With the increasing solar penetration rate,the operation stability of photovoltaic power plants has an important impact on the safe operation of the power system.Therefore,the State Grid Corporation requires that the photovoltaic power generation system have a certain low voltage ride through capability.The faults it targets usually refer to grid short-circuit faults However,the most common short-circuit fault is the asymmetric fault,so it is of practical significance to study the control strategy of the MMC photovoltaic grid-connected system under the grid-side asymmetric fault.In this paper,the MMC photovoltaic grid-connected system is taken as the research object,the MMC operating conditions under the asymmetric fault condition of the power grid are studied,and the control strategy is formulated in the AC-side inverter part and the DC-side photovoltaic power generation unit part using model predictive control..The steady-state operation of the modular multilevel converter needs to achieve AC current control,phase-to-phase circulating current suppression,sub-module capacitor voltage balance and other goals.Model predictive control is suitable for this multi-variable,multi-constrained nonlinear system.When an asymmetric fault occurs on the side,the relevant control mechanism will be different from the normal operating conditions,and specific analysis is required.First introduced the topology structure of MMC photovoltaic grid-connected system,discussed the operation principle of MMC,introduced the method to realize the maximum power tracking control of photovoltaic power generation unit,on this basis,the continuous mathematical model and discrete mathematical model on the MMC AC side were derived,for the follow-up Model predictive control makes theoretical preparations.The effect of different control targets on the AC side on the performance of MMC capacitor voltage fluctuations and AC current peaks under asymmetric faults on the grid side is analyzed in detail,and the elimination of the negative sequence component of AC current is determined as the control target.In addition,the positive and negative sequence separation of the asymmetric AC quantity is very important to calculate the AC current reference value in the control process.The second-order generalized integrator is used to achieve the positive and negative sequence separation in the two-phase stationary coordinate system.The basic principle of model predictive control is introduced,and the idea of hierarchical predictive control is used to achieve the control goals of AC current tracking and circulating current suppression,respectively.In AC current control,considering the imbalance of power will increase the risk of peak output current,a strategy to limit the peak has been formulated.In the circulating current suppression link,firstly,the change of the bridge arm power under an asymmetric fault is analyzed.On this basis,the circulating current component is deduced in detail,and the traditional single-order predictive control method is improved.For notable.In valve-level control,improving the traditional capacitor voltage sequencing method can not only ensure the balance of capacitor voltage,but also reduce the switching frequency of the device,reduce switching losses,and extend the service life of the device.Considering that suppressing the peak value of AC current on the AC side will reduce the active power of the MMC output,in order to ensure the power balance on the AC and DC sides,constant power control is implemented through model predictive control to reduce the active power output of the photovoltaic unit and ensure the stability of the DC bus voltage.Establish the discrete transfer function of the AC side model predictive control algorithm,use the Juli criterion to verify the stability of the algorithm,and make corresponding improvements to the algorithm to improve its stability margin.When the inductance and other parameters change,the system can continue to maintain stable operation.Finally,a model of the MMC photovoltaic grid-connected system was built on the RT-LAB real-time simulation platform.By dividing the model,compiling and running,the above control theory was verified and analyzed.Compared with the traditional PR control strategy,the model predictive control has shorter response time and better dynamic performance.