Efficiency optimization control of permanent magnet synchronous motors for pure electric vehicles based on GBDT


Vol. 24, No. 2, pp. 215-226, Feb. 2024
10.1007/s43236-023-00716-x




 Abstract

In this paper, “Machine learning” is introduced to motor efficiency optimization control to improve the operating efficiency of the permanent magnet synchronous motors (PMSMs) for pure electric vehicles. A current distribution method based on the gradient boosting decision tree (GBDT) is proposed. The efficiency of the motor operation can be improved by coordinating the current control. First, a mathematical model of the motor efficiency is established, and the current distribution law of the optimal efficiency of the motor in different operating regions is qualitatively analyzed. The control system is based on this current distribution. Second, the sample space is established based on measured data, where the current regression model of the GBDT is introduced. Then by analyzing the importance of characteristic variables, the structure of the model is optimized, and the input and output of the model are reasonably selected, which are embedded into the control system to realize the coordinated control of the current. Finally, comparative experiment shows that the proposed method can improve the efficiency of PMSMs in the whole speed range.


 Statistics
Show / Hide Statistics

Cumulative Counts from September 30th, 2019
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.



Cite this article

[IEEE Style]

F. Xie, H. Wang, S. Ni, C. An, "Efficiency optimization control of permanent magnet synchronous motors for pure electric vehicles based on GBDT," Journal of Power Electronics, vol. 24, no. 2, pp. 215-226, 2024. DOI: 10.1007/s43236-023-00716-x.

[ACM Style]

Fang Xie, Houying Wang, Shilin Ni, and Chaochen An. 2024. Efficiency optimization control of permanent magnet synchronous motors for pure electric vehicles based on GBDT. Journal of Power Electronics, 24, 2, (2024), 215-226. DOI: 10.1007/s43236-023-00716-x.