Model Predictive Control And Grid-Connected Voltage Governance Efficiency Evaluation For Cascaded H-Bridge Power Conversion System

The voltage quality of the grid-connected point is easily impacted by a three-phase imbalance or distortion disturbances such as distributed generation,asymmetric load,and nonlinear load in the modern distributed power grid with high penetration of clean energy and a high proportion of power electronic devices.Power electronic grid-connected equipment like distributed power inverters and power conversion systems are significantly impacted by the grid-connected voltage conditions in terms of their performance and output power quality.The distributed power grid-connected inverter can utilize its residual capacity to both realize the voltage quality control function of the grid-connected point as well as the traditional grid-connected function through reasonable control methods.Consequently,the following topics are the main focus of the thesis.First,the traditional power quality issues are discussed from both the steady-state and dynamic perspectives.Following that,a major focus is placed on the typical power quality issues brought on by the widespread usage of distributed generation and nonlinear load access in the modern distributed power grid.Discussed are the key risks posed by distorted and imbalanced non-ideal grid voltage situations.Secondly,the non-isolated bidirectional half-bridge DC/DC converter is used as the pre-stage DC converter topology of the power conversion system,and the cascaded H-bridge converter is used as the post-stage DC / AC topology.Through the modeling and discretization of the cascaded H-bridge power conversion system,the predicted value of the current at the next moment of the converter is calculated.The switching state of the power electronic device in the converter at the next moment is obtained by the objective function,and the model predictive control of the power conversion system under ideal grid conditions is realized.Aiming at the non-ideal grid conditions,the p-q-r method is used to obtain the unbalanced and distorted current components that need to be compensated,which are used as the reference current of model predictive control.The model predictive control of the power conversion system under non-ideal grid conditions is realized,which can ensure a good grid-connected effect and improve the voltage quality of grid-connected point.Thirdly,considering the practical application scenario of a cascaded H-bridge power conversion system in a medium-voltage microgrid cluster,the capacity of the power conversion system is divided,part of which is used to control power quality,and the remaining capacity is connected to the grid normally.Aiming at the problem of limited compensation capacity of converters,a power coordinated control method between converters based on model predictive control under non-ideal grid conditions is further studied.Under the condition of asymmetric load,both the cascaded H-bridge power conversion system and three-level NPC converter show a good three-phase unbalance suppression effects.The power coordinated control between them can effectively improve the three-phase unbalance suppression ability of the whole system.The control method studied can improve the system voltage quality in multiple scenarios.Finally,the thesis takes the cascaded H-bridge power conversion system and the three-level NPC converter under different compensation capacities as the object and takes the power quality data of the two converters after the voltage control of each node in the microgrid cluster as the sample,and establishes a super-efficiency SBM model.Based on the model,the grid-connected voltage control efficiency of different converters with asymmetric load and nonlinear load under different compensation capacity settings is studied.