Search

This paper proposes an equivalent temperature estimation‐based finite control set model predictive control (ETE‐FCS‐MPC) method to balance the thermal stress among transistors in three level neutral point clamp (3L‐NPC) traction inverters, which can be used to prolong the whole service life and improve the reliability of the traction inverter. Compared with other methods, the proposed ETE‐FCS‐MPC method requires lesser calculation without additional sensor. Firstly, an energy loss‐based equivalent temperature estimation method is proposed to avoid the complexity of junction temperature prediction. Secondly, an energy variance‐based thermal stress balance control objective is introduced. Then, a normalized evaluation function is established to select the optimal switching state. Finally, the effectiveness of this method is verified on hardware‐in‐the‐loop tests.

The three-level neutral point clamped (3LNPC) inverter has the problem of unbalanced power loss and temperature. This paper proposes a temperature balance method based on finite control set modulation model predictive control (FCS$-\mathrm{M}^{2}$ PC). Firstly, the periodic current of the current switch is measured and the delay compensation is made to predict the current and power loss of the next switching cycle. Then, the evaluation function value is calculated to find the optimal switch state sequence and the switching time. The output is the control signal that instructs the switch to move in sequence at the corresponding time. After the current switch sequence is complete, the prediction of the next switch cycle begins. This method can be used in low and medium switching frequency systems, and it does not need to make prediction in every sampling period, which saves computation. The simulation results show the effectiveness of the proposed method.

The there-level Neutral Point Clamped(3L-NPC) inverter is widely used in various applications. One of shortcoming in there-level NPC is the uneven temperature distribution caused by uneven power loss distribution. For this reason, this paper proposes a temperature balance control (TBC) method based on finite control set model predictive control (FCS-MPC) to improve the temperature distribution of inverters. Firstly, the evaluation function of the FCS-MPC is established, which includes current tracking, power tracking and current, power limiting. Based on the discrete model, a predicted value of the three-phase current and transistor power loss corresponding to all active switch states in the next sampling period is then calculated. Finally, in all the effective switch states, a switch state is selected, which corresponds to the smallest evaluation function value. The switch state will be used as the control signal for the next sampling period. The simulation results show that the proposed TBC method can not only balance the temperature distribution, but also has little effect on the system control performance.