Model Predictive Direct Power Control Based On Improved T-Type Inverter

At present,all countries in the world are vigorously developing new energy technologies.Among them,the development of wind power technology is relatively mature and has been widely used in the industrial field.The output of wind power generation is mostly high-power,so most of the wind power converters are multi-level converter topologies,and most of them are three-level converters,so the performance of the three-level inverter is very essential in the wind power converter group.In this paper,the improved T-type three-level inverter topology is used,and the model predictive direct power control of the inverter is studied based on this topology.Firstly,it introduces the current development status of wind power technology at home and abroad,the development history of three-level inverters,the current mainstream three-level inverter topology and the research status of inverter control strategies.Which make the application background more clearly.After analyzing the performance of several mainstream three-level inverter topologies,The improved T-type three-level inverter topology is selected.In addition to inheriting the advantages of the traditional T-type three-level inverter,it is also significant that reducing the conduction loss in the zero state,thereby improving the inverter output efficiency.Secondly,the model predictive control algorithm is briefly introduced in this paper,and the specific implementation method of the algorithm based the three-level inverter is analyzed.In the dq rotating coordinate system,a control method combining model predictive control algorithm and direct power control algorithm is proposed,which is model predictive direct power control,and Mathematical modeling and theoretical analysis of improved T-type three-level inverter based the model predictive direct power control method.The control divides the 27 switching states into two groups,and compares the two capacitor voltages on the DC side to determine which group of switching states will be optimized for rolling,selecting the optimal switching state at the current time,and thus implementing the inverter.control.The group control theory of the model predicting direct power control in the dq coordinate system not only ensures the balance of theneutral-point voltage,but also reduces the number of rolling optimizations,thereby reducing the complexity of the algorithm and reducing the computational burden.At the same time,the model is established in the dq coordinate system,which realizes the decoupling of active power and reactive power,and improves the flexibility of control.Compared with the traditional multi-level model predictive control,the method does not need to predict the neutral point voltage trend,and does not need to perform the rolling optimization of the neutral-point voltage in the cost function,avoiding the design of the correlation weighting factor,and the control method is more direct and effective.Finally,based on the theoretical analysis and simulation results,the experimental platform was constructed and the related control program was written,which verified the model predictive direct power control.Results show that the model predictive direct power control based on improved T-type three-level inverter topology can improve system robustness,fast dynamic response,and stability.