
浏览全部资源
扫码关注微信
1.辽宁省流程工业节能与绿色低碳技术工程研究中心 东北大学 沈阳 110819
2. 中国建筑科学研究院 北京 100013
3. 北京理工大学机械与车辆学院 北京 100081
韩宗伟,男,教授,东北大学冶金学院,15040168696,E-mail:hanzongwei_neu@163.com。研究方向:高精度控温技术,绿色/高效制冷相关理论及其关键技术。
收稿:2025-06-02,
修回:2025-06-10,
录用:2025-07-15,
纸质出版:2026-02-16
移动端阅览
张义奇,黄烁全,历秀明等.热管/蒸气压缩复合空调系统故障诊断模型分类解释性研究[J].制冷学报,2026,47(01):88-95.
Zhang Yiqi,Huang Shuoquan,Li Xiuming,et al.Interpretability Study on the Fault-Diagnosis Model of the Heat Pipe / Vapor-Compression Composite Air-Conditioning System[J].Journal of Refrigeration,2026,47(01):88-95.
张义奇,黄烁全,历秀明等.热管/蒸气压缩复合空调系统故障诊断模型分类解释性研究[J].制冷学报,2026,47(01):88-95. DOI: 10.12465/issn.0253-4339.20250602001. CSTR: XXXXX.XX.XXX.20250602001.
Zhang Yiqi,Huang Shuoquan,Li Xiuming,et al.Interpretability Study on the Fault-Diagnosis Model of the Heat Pipe / Vapor-Compression Composite Air-Conditioning System[J].Journal of Refrigeration,2026,47(01):88-95. DOI: 10.12465/issn.0253-4339.20250602001. CSTR: XXXXX.XX.XXX.20250602001.
将数据驱动的故障诊断模型用于数据中心空调系统,可有效提高其运行可靠性。但此类模型通常缺乏诊断依据,限制了其广泛应用。本文建立了基于典型机器学习算法的复合空调系统故障诊断模型,对比了各模型诊断性能,并基于SHAP(shapley additive explanation)方法对诊断模型进行了可解释性研究。结果表明:基于卷积神经网络(convolutional neural network,CNN)的故障诊断模型在热管及蒸气压缩模式下性能均为最优,在各分类下F-1值均高于0.999。热管模式下,CNN模型诊断所依据的主要特征为冷凝器风机频率、室外温度及制冷剂泵功耗;在蒸气压缩模式下则为室外温度、压缩机频率和过冷度。
Applying data-driven fault-diagnosis models to data center air-conditioning systems can significantly improve operational reliability. However, these models often lack diagnostic interpretability, which limits their application. This study develops a composite fault-diagnosis model based on typical machine-learning algorithms, compares the diagnostic performance of different models, and conducts interpretability research on the diagnostic models using the Shapley additive explanation method. The results demonstrate that the convolutional neural network (CNN)-based fault-diagnosis model achieves optimal performance in both the heat-pipe and vapor-compression modes, with F-1 scores exceeding 0.999 across all classifications. In the heat-pipe mode, the diagnosis of the CNN model primarily relies on the condenser-fan frequency, outdoor temperature, and refrigerant-pump power consumption as key features, whereas in the vapor-compression mode, the dominant features are the outdoor temperature, compressor frequency, and subcooling degree.
陈瑞 , 许家翔 , 曹军 . 数据中心间接蒸发冷却复合空调系统的节能运行分析 [J]. 暖通空调 , 2024 , 54 ( 10 ): 113 - 119 .
CHEN Rui , XU Jiaxiang , CAO Jun . Energy saving operation analysis of indirect evaporative cooling composite air conditioning systems in data centers [J]. Journal of HV & AC , 2024 , 54 ( 10 ): 113 - 119 .
许梦玫 , 翟晓强 , 李国柱 , 等 . 数据中心冷却技术的研究进展 [J]. 建筑科学 , 2018 , 34 ( 8 ): 124 - 132 .
XU Mengmei , ZHAI Xiaoqiang , LI Guozhu , et al . Research progress in cooling technology of data centers [J]. Building Science , 2018 , 34 ( 8 ): 124 - 132 .
陈心拓 , 周黎旸 , 张程宾 , 等 . 绿色高能效数据中心散热冷却技术研究现状及发展趋势 [J]. 中国工程科学 , 2022 , 24 ( 4 ): 94 - 104 .
CHEN Xintuo , ZHOU Liyang , ZHANG Chengbin , et al . Research status and future development of cooling technologies for green and energy-efficient data centers [J]. Strategic Study of CAE , 2022 , 24 ( 4 ): 94 - 104 .
张海南 , 邵双全 , 田长青 . 数据中心自然冷却技术研究进展 [J]. 制冷学报 , 2016 , 37 ( 4 ): 46 - 57 .
ZHANG Hainan , SHAO Shuangquan , TIAN Changqing . Research advances in free cooling technology of data centers [J]. Journal of Refrigeration , 2016 , 37 ( 4 ): 46 - 57 .
吕继祥 , 王铁军 , 赵丽 , 等 . 基于自然冷却技术应用的数据中心空调节能分析 [J]. 制冷学报 , 2016 , 37 ( 3 ): 113 - 118 .
LYU Jixiang , WANG Tiejun , ZHAO Li , et al . Energy saving analysis of data center air conditioning system based on application of natural cooling technology [J]. Journal of Refrigeration , 2016 , 37 ( 3 ): 113 - 118 .
褚俊杰 , 徐伟 , 霍慧敏 . 数据中心间接蒸发自然冷却空调系统在中国的适用性分析 [J]. 制冷与空调(四川) , 2021 , 35 ( 5 ): 725 - 732 .
CHU Junjie , XU Wei , HUO Huimin . Applicability analysis of indirect free cooling air conditioning system for China data center [J]. Refrigeration & Air Conditioning , 2021 , 35 ( 5 ): 725 - 732 .
ZOU Sikai , ZHANG Quan , YU Yuebin , et al . Field study on the self-adaptive capacity of multi-split heat pipe system (MSHPS) under non-uniform conditions in data center [J]. Applied Thermal Engineering , 2019 , 160 : 113999 .
王铁军 , 王飞 , 李宏洋 , 等 . 动力型分离式热管设计与试验研究 [J]. 制冷与空调(北京) , 2014 , 14 ( 12 ): 41 - 43 .
WANG Tiejun , WANG Fei , LI Hongyang , et al . Design and experimental study of dynamic separate type heat pipe [J]. Refrigeration and Air-conditioning , 2014 , 14 ( 12 ): 41 - 43 .
王飞 , 邵双全 , 张海南 . 数据中心冷却用动力型热管的实验研究 [J]. 制冷学报 , 2020 , 41 ( 4 ): 89 - 96 .
WANG Fei , SHAO Shuangquan , ZHANG Hainan . Experimental study on compressor-driven loop heat pipe for data center cooling [J]. Journal of Refrigeration , 2020 , 41 ( 4 ): 89 - 96 .
HAN Zongwei , ZHANG Yanqing , MENG Xin , et al . Simulation study on the operating characteristics of the heat pipe for combined evaporative cooling of computer room air-conditioning system [J]. Energy , 2016 , 98 : 15 - 25 .
黄倩云 , 陈焕新 , 孙劭波 , 等 . 基于支持向量机的多联机系统制冷剂充注量故障检测与诊断 [J]. 暖通空调 , 2018 , 48 ( 1 ): 91 - 95 .
HUANG Qianyun , CHEN Huanxin , SUN Shaobo , et al . SVM-based FDD method for refrigerant charge in variable refrigerant flow system [J]. Journal of HV & AC , 2018 , 48 ( 1 ): 91 - 95 .
GUO Yabin , TAN Zehan , CHEN Huanxin , et al . Deep learning-based fault diagnosis of variable refrigerant flow air-conditioning system for building energy saving [J]. Applied Energy , 2018 , 225 : 732 - 745 .
王路瑶 , 吴斌 , 杜志敏 , 等 . 基于长短期记忆神经网络的数据中心空调系统传感器故障诊断 [J]. 化工学报 , 2018 , 69 ( 增刊2 ): 252 - 259 .
WANG Luyao , WU Bin , DU Zhimin , et al . Sensor fault detection and diagnosis for data center air conditioning system based on LSTM neural network [J]. CIESC Journal , 2018 , 69 ( Suppl.2 ): 252 - 259 .
ZHOU Zhenxin , LI Guannan , CHEN Huanxin , et al . Fault diagnosis method for building VRF system based on convolutional neural network: considering system defrosting process and sensor fault coupling [J]. Building and Environment , 2021 , 195 : 107775 .
LI Guannan , YAO Qing , FAN Cheng , et al . An explainable one-dimensional convolutional neural networks based fault diagnosis method for building heating, ventilation and air conditioning systems [J]. Building and Environment , 2021 , 203 : 108057 .
熊成龙 , 李冠男 , 劳春峰 , 等 . 卷积神经网络的空调系统故障诊断可解释研究 [J]. 家电科技 , 2024 ( 增刊1 ): 170 - 174 .
XIONG Chenglong , LI Guannan , LAO Chunfeng , et al . Interpretation study on convolutional neural networks-based fault diagnosis of air conditioning system [J]. Journal of Appliance Science & Technology , 2024 ( Suppl.1 ): 170 - 174 .
DU Zhimin , CHEN Kang , CHEN Siliang , et al . Deep learning GAN-based data generation and fault diagnosis in the data center HVAC system [J]. Energy and Buildings , 2023 , 289 : 113072 .
ZHANG Yiqi , LI Mengyi , DONG Jiaxiang , et al . Study on the impacts of refrigerant leakage on the performance and reliability of datacenter composite air conditioning system [J]. Energy , 2023 , 284 : 129336 .
LI Xiuming , ZHANG Ce , DONG Jiaxiang , et al . Feasibility investigation on a novel rack-level cooling system for energy-saving retrofit of medium-and-small data centers [J]. Applied Thermal Engineering , 2023 , 229 : 120644 .
PANG K . A comparative study of explainable machine learning models with Shapley values for diabetes prediction [J]. Healthcare Analytics , 2025 , 7 : 100390 .
0
浏览量
430
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