
浏览全部资源
扫码关注微信
1.天津商业大学 天津市制冷技术重点实验室,天津 300134
2. 哈尔滨工业大学电气工程及自动化学院,哈尔滨 150006
刘圣春,男,教授,天津商业大学机械工程学院,13920682426,E-mail:liushch@tjcu.edu.cn。研究方向:CO2制冷热泵,制冷系统优化及节能技术,冷冻冷藏技术,相变储能及新能源利用。
收稿:2026-06-01,
修回:2026-06-22,
录用:2026-06-25,
网络首发:2026-08-17,
移动端阅览
李雪强,张植青,张一凡,等. 应用于电机散热的热管性能特性及预测模型对比[J]. 制冷学报,XXXX,XX(XX):1-8.
Li Xueqing,Zhang Zhiqing,Zhang Yifan,et al. Heat Pipe Performance Characteristics and Prediction Models for Motor Heat Dissipation[J]. Journal of Refrigeration,XXXX,XX(XX):1-8.
李雪强,张植青,张一凡,等. 应用于电机散热的热管性能特性及预测模型对比[J]. 制冷学报,XXXX,XX(XX):1-8. DOI: 10.12465/issn.0253-4339.20260601001.
Li Xueqing,Zhang Zhiqing,Zhang Yifan,et al. Heat Pipe Performance Characteristics and Prediction Models for Motor Heat Dissipation[J]. Journal of Refrigeration,XXXX,XX(XX):1-8. DOI: 10.12465/issn.0253-4339.20260601001.
电机高功率密度化使其热管理问题日趋严峻,准确预测热管等效导热系数对电机散热设计至关重要。针对电机散热系统中热管性能及性能预测问题,本文以等效导热系数为评价指标,实验研究了热源功率、冷凝段与蒸发段的相对位置、弯折角度等关键参数对热管传热性能的影响机制。在此基础上建立并对比了遗传算法优化的反向传播神经网络(GA-BPNN)、卷积神经网络(CNN)、最小二乘支持向量机(LSSVM)、随机森林(RF)与极端梯度提升(XGBoost)5种机器学习模型的性能。结果表明:重力与弯折角度对热管的导热性能影响最大,上方侧90°弯折的热管导热性能最好,其导热性能是下方侧的30倍;XGBoost模型的平均绝对百分比误差为3.72%,决定系数为0.95,均方根误差仅次于GA-BPNN模型,在所选的5种模型中的精度最高,且训练时间小于2 s;XGBoost模型的预测表明不同安装位置及弯折角度的热管性能不尽相同,为了提高电机各绕组的均温性,推荐弯折角度为0°的热管用于电机散热中。
Electric motors are developing towards high power density and miniaturization, making thermal management increasingly critical. This study investigated the influence mechanisms of key parameters, such as the heat source power, position between evaporator and condenser side, and bending angle, on the heat transfer performance of heat pipe. Moreover, comparative analyses were performed on the prediction performance of five machine learning models: genetic algorithm optimized back propagation neural network (GA-BPNN), convolutional neural network (CNN), least squares support vector machine (LSSVM), random forest (RF), and extreme gradient boosting (XGBoost). The results indicate that: the coupled detrimental effects of gravity and bending angle on the capillary wick structure govern the heat transfer performance; under a 90° bending angle at 60-W heat source power, the side arrangement of the condenser above evaporator achieves the best equivalent thermal conductivity, which is 30 times that of the condenser under evaporator. The mean absolute percentage error of XGBoost is 3.72%, coefficient of determination is 0.95, root mean square error only higher than that of the GA-BPNN model, and training time is less than 2 s, demonstrating the best performance among five models. The prediction results of XGBoost model illustrate that different positions and bending angles lead to different dissipation performances. To improve the temperature uniformity in the winding of motor, a heat pipe with 0° bending angle is recommended for the heat dissipation of motor.
Wang Jinghui , Rakha H A . Electric train energy consumption modeling [J]. Applied Energy , 2017 , 193 : 346 - 355 .
江从喜 , 赵兰萍 , 杜旭之 , 等 . 基于整车工况的电动汽车轮毂电机散热分析 [J]. 中国机械工程 , 2016 , 27 ( 13 ): 1839 - 1845 .
Jiang Congxi , Zhao Lanping , Du Xuzhi , et al . Thermal analysis on in-wheel motor under whole electric vehicle driving conditions [J]. China Mechanical Engineering , 2016 , 27 ( 13 ): 1839 - 1845 .
毛正松 , 陆永卷 , 欧阳石坤 , 等 . 增程器风冷电机散热方案仿真分析及测试验证 [J]. 内燃机与配件 , 2023 ( 5 ): 27 - 29 .
Mao Zhengsong , Lu Yongjuan , Ouyang Shikun , et al . Simulation analysis and test verification of an air cooled E-motor heat dissipation design for range extender [J]. Internal Combustion Engine & Parts , 2023 ( 5 ): 27 - 29 .
陈前 , 周亚南 , 徐高红 , 等 . 混合转子永磁电机双风道冷却系统设计 [J]. 电气工程学报 , 2024 , 19 ( 2 ): 1 - 8 .
Chen Qian , Zhou Yanan , Xu Gaohong , et al . Design of dual air duct cooling system for hybrid rotor permanent magnet motor [J]. Journal of Electrical Engineering , 2024 , 19 ( 2 ): 1 - 8 .
张良 , 陈克 , 张得胜 , 等 . 纯电动巴士电机驱动器冷却系统的设计和分析 [C]// 2014中国汽车工程学会年会论文集 . 2014 : 93 - 95 .
Zhang Liang , Chen Ke , Zhang Desheng , et al . Design and analysis of cooling system for motor drive of pure electric bus [C]// Proceedings of 2014 SAE China Annual Conference . 2014 : 93 - 95 .
Li Chenxi , Li Ji . Passive cooling solutions for high power server CPUs with pulsating heat pipe technology: review [J]. Frontiers in Energy Research , 2021 , 9 : 755019 .
Liu Chuan , Xu Zeyuan , Gerada D , et al . Experimental investigation on oil spray cooling with hairpin windings [J]. IEEE Transactions on Industrial Electronics , 2020 , 67 ( 9 ): 7343 - 7353 .
Li Ye , Li Qi , Fan Tao , et al . Heat dissipation design of end winding of permanent magnet synchronous motor for electric vehicle [J]. Energy Reports , 2023 , 9 : 282 - 288 .
Wang Huimin , Zhang Chujie , Guo Liyan , et al . Novel revolving heat pipe cooling structure of permanent magnet synchronous motor for electric vehicle [J]. Applied Thermal Engineering , 2024 , 236 : 121641 .
Chen Zutao , Yu Zhongjun , Fu Jia , et al . Analysis and design of air-heat pipe composite cooling of high power density motor [J]. Applied Thermal Engineering , 2024 , 236 : 121495 .
Alammar A A , Al-Dadah R K , Mahmoud S M . Numerical investigation of effect of fill ratio and inclination angle on a thermosiphon heat pipe thermal performance [J]. Applied Thermal Engineering , 2016 , 108 : 1055 - 1065 .
田金颖 , 诸凯 , 刘建林 , 等 . 冷却电子芯片的平板热管散热器传热性能研究 [J]. 制冷学报 , 2007 , 28 ( 6 ): 18 - 22 .
Tian Jinying , Zhu Kai , Liu Jianlin , et al . Experiments investigation on thermal conductivity of flat heat pipes used to cool electric die [J]. Journal of Refrigeration , 2007 , 28 ( 6 ): 18 - 22 .
赵萧涵 , 刘圣春 , 李雪强 , 等 . U型重力热管启动过程中压力和温度的演变规律 [J]. 制冷学报 , 2024 , 45 ( 6 ): 106 - 113 .
Zhao Xiaohan , Liu Shengchun , Li Xueqiang , et al . Evolution of pressure and temperature during start-up process in U-shaped gravity heat pipe [J]. Journal of Refrigeration , 2024 , 45 ( 6 ): 106 - 113 .
Wansasueb K , Promthaisong P , Bubphachot B , et al . Applied machine learning: Performance prediction of heat pipe with mesh wick [J]. Case Studies in Thermal Engineering , 2024 , 63 : 105307 .
Jin I J , Park Y Y , Bang I C . Heat transfer performance prediction for heat pipe using deep learning based on wick type [J]. International Journal of Thermal Sciences , 2024 , 197 : 108806 .
Maddah H , Ghazvini M , Ahmadi M H . Predicting the efficiency of CuO/water nanofluid in heat pipe heat exchanger using neural network [J]. International Communications in Heat and Mass Transfer , 2019 , 104 : 33 - 40 .
Huang Guanghan , Li Wenming , Zhong Guisheng , et al . Optimizing L-shaped heat pipes with partially-hybrid mesh-groove wicking structures [J]. International Journal of Heat and Mass Transfer , 2021 , 170 : 120926 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621