基于全局热网络和局部流体动力学建模的全封闭永磁牵引电机热性能优化（英文）

Thermal optimization of a totally enclosed forced ventilated permanent magnet traction motor using lumped parameter and partial computational fluid dynamics modeling

目的:提出一种适用于全封闭冷却结构的电机热性能优化模型,设计一台600 kW的高速列车用永磁牵引电机。创新点:1.通过耦合局部流体动力学模型的方法求解电机复杂冷却风道内的对流传热系数,并在全局热网络模型的框架内得到快速、准确的电机温升结果以用于结构优化;2.在冷却风道中引入栅格结构,采用热性能分析模型优化冷却结构,提升电机热性能;3.通过三维流体动力学模型计算电机局部温升最大值,并提山一种预测特定结构下电机铁损工作阈值的工程方法。方法:1.采用热网络法建立全局热网络模型(图3),并通过耦合局部流体动力学模型计算风道内的热网络参数(图4和6); 2.应用田口设计法对电机风道结构进行优化,并研制样机进行验证(计算与试验结果见表5); 3.假设铁损的谐波附加值与磁密值成正比,通过三维流体动力学模型计算给山端部绕组、永磁体温升值与铁损的预测曲线,并用样机试验进行验证。结论:1.采用全局热网络和局部流体动力学建模的方法可以快速、正确地计算复杂冷却结构下的电机温升分布,且优化后的冷却结构至少可以提升文中电机15%的热性能;2.本文提出的优化模型适用于全封闭风冷或者水冷等冷却结构相对独立且尚无经验公式可参考的电机热性能优化设计;3.铁损工作阈值的预测方法可以为电磁和控制系统设计提供参考。

In this study, we present a thermal optimization method using the overall lumped parameter（LP） and partial computational fluid dynamics（CFD） modeling for a 600-kW permanent magnet traction motor developed for high-speed trains. The motor is totally enclosed forced ventilated to achieve high power density, high efficiency, and low maintenance requirements. Considering the electro-magnetic performance, bogie space, and thermal capacity,we propose a ventilation structure with zigzag plates in sector cross-section. We focus particularly on the ventilation channels and propose an overall LP model for thermal optimization, in which the full consideration of the influence of turbulent flow is given by using a partial CFD model. Given the specific critical parameters from the optimization results, we present a complete 3 D CFD model of the whole motor to obtain an accurate temperature distribution and the maximum temperature rises in local points. The benefit of zigzag plates is studied extensively using both the LP and the complete CFD models and the results are verified by equivalent thermal experiments under rated operations. Experimental results indicate that the ventilation structure fulfills the normal operational demands of high-speed trains by improving thermal performance by more than 15%. Additionally, we propose an engineering method to estimate iron loss constraint with the complete CFD model to guide the control system design.