پیش بینی کنترل قدرت مستقیم بر پایه مبدل متصل به شبکه نوع T
Abstract
I- Introduction
II- Improved t-type three-level inverter
III- Grid system mathematical model
IV- Model predictive direct power control strategy
V- Simulation and experiment
VI- Conclusion
References
Abstract
In this paper, the improved T-type three-level topology can significantly reduce the conduction loss in zero states compared with the traditional T-type three-level topology. In the DQ rotating coordinate system, a control method combining the model predictive algorithm and the direct power control is proposed. This control method divides 27 kinds of switch states into two groupsand decides which group of switch states is to be scrolled optimally by comparing the two capacitor voltages on the DC side. This method reduces the computational burden by reducing the number of rolling optimization under the condition of ensuring the neutral point voltage balance. Compared with the traditional multilevel model predictive control, the method eliminates the need of predicting the neutral point voltage as well as the cost function computation, and the control method is more direct and effective. Simulation and experimental results show that the proposed model predictive direct power control (MPDPC) method based on the improved T-topology can improve system robustness, fast dynamic response, wide operating range and stability.
Introduction
The inverter is a key part of the grid-connected system. It has high requirements on the quality and conversion efficiency of the waveform. A control method with good performance is very important. Improved T-type three-level inverter not only inherits the advantages of the traditional T-type three-level converter such as less component count,no clamping diodes through the current path,but also has more excellent performance in conduction loss. Compared with the traditional T-type three-level inverter topology, the improved T-type inverter reduces conduction losses by 90% under zero-vector conditions, which is equivalent to a total loss reduction of 30% [1]. MPC (Model Predictive Control) has the advantages of fast dynamic response, strong anti-interference ability, the low requirement on the model and the obvious advantages in dealing with nonlinear systems and multi-control targets. Ever since its inception, it has been constantly welcomed by industries system with strong disturbances and time-varying large inertia. It has been widely used in power electronics [2], motor-driven [3], solar wind power conversion [4],[5] and another control system. However, the traditional model predictive control increases dramatically as the number of levels increases and the number of control objectives increases. Direct power control has the advantages of good robustness, fast dynamic response, and a simple control structure. It has been widely used in practice. In this paper, the combination of model predictive control and direct power control not only inherits the advantages of good robustness of direct power control and simple control structure but also achieves the control of neutral point balance and constant power output.