Automatic dryness judgment method for heat pump clothes dryers based on multi-dimensional time series feature engineering and ensemble learning

Liao Min; Zhang Shurong; Zhao Jiajun

doi:10.12465/issn.0253-4339.20260508003

您当前的位置：

首页 >

文章列表页 >

Automatic dryness judgment method for heat pump clothes dryers based on multi-dimensional time series feature engineering and ensemble learning

更新时间：2026-08-17

- Automatic dryness judgment method for heat pump clothes dryers based on multi-dimensional time series feature engineering and ensemble learning
- Journal of Refrigeration Pages: 1-11(2026)
- 作者机构：
  
  小米科技（武汉）有限公司武汉 430010
- 作者简介：
  
  Liao Min， male， senior engineer， Xiaomi Technology （Wuhan）Co.， Ltd.， 86-18578272392， E-mail： liaomin1@xiaomi.com. Research fields： intelligent monitoring and control of household appliances， design and development of heat pump systems.
- 基金信息：
- DOI：10.12465/issn.0253-4339.20260508003
  CLC： TM925.34;TP273
- Received：08 May 2026，
  
  Revised：2026-06-18，
  
  Accepted：25 June 2026，
  
  Online First：17 August 2026，
- 稿件说明：
移动端阅览
廖敏，刘运涛，杜国庆，等. 基于多维时序特征工程与集成学习的热泵干衣机自动判干方法［J］. 制冷学报，XXXX，XX（XX）：1-11.

Liao Min，Liu Yuntao，Du Guoqing，et al. Automatic dryness judgment method for heat pump clothes dryers based on multi-dimensional time series feature engineering and ensemble learning［J］. Journal of Refrigeration，XXXX，XX（XX）：1-11.
廖敏，刘运涛，杜国庆，等. 基于多维时序特征工程与集成学习的热泵干衣机自动判干方法［J］. 制冷学报，XXXX，XX（XX）：1-11. DOI： 10.12465/issn.0253-4339.20260508003.

Liao Min，Liu Yuntao，Du Guoqing，et al. Automatic dryness judgment method for heat pump clothes dryers based on multi-dimensional time series feature engineering and ensemble learning［J］. Journal of Refrigeration，XXXX，XX（XX）：1-11. DOI： 10.12465/issn.0253-4339.20260508003.

摘要

针对现有热泵干衣机规则式阈值判干策略存在烘干时间冗余和负载过烘的问题，本文提出一种基于多维时序特征工程与集成学习的自动判干方法。该方法基于热泵系统运行机理提取多维时序特征，结合弱监督标签生成与“标签纯度筛选-折外交叉预测”双重样本清洗策略，构建融合XGBoost、LightGBM和CatBoost的集成学习模型，并引入双阈值置信度锁以降低“湿判干”风险。结果表明：该模型在静态数据集上的测试准确率为91.26%，F1分数为92.17%，引入置信度锁后高置信度样本判定准确率提升至96.30%；在IEC负载和实物负载验证测试中，相比规则式阈值判干方案烘干时间分别缩短30.9%和22.4%，且均满足判干要求。所提方法可有效缓解烘干时间冗余和过烘问题，具有较好的工程应用价值。

Abstract

To address the redundant drying time and over-drying problems of rule-based threshold strategies in heat pump clothes dryers， this paper proposes an automatic dryness judgment method based on multi-dimensional time series feature engineering and ensemble learning. The multi-dimensional time series features were extracted according to the operating characteristics of heat pump systems. By combining weakly supervised label generation with a dual sample-cleaning strategy involving label-purity filtering and out-of-fold prediction， an ensemble learning model integrating XGBoost， LightGBM， and CatBoost was established， and a dual-threshold confidence lock was further introduced to reduce the risk of false dry judgments. The results show that the proposed model achieved a test accuracy of 91.26% and F1 score of 92.17% on the static dataset， and the judgment accuracy of high-confidence samples increased to 96.30% after applying the confidence lock. In International Electrotechnical Commission-load and real-load validation tests， the drying time was reduced by 30.9% and 22.4%， respectively， compared with the rule-based threshold strategy while still satisfying the drying requirements. The proposed method can effectively alleviate redundant drying time and over-drying problems and shows good engineering application potential.

关键词

Keywords

references

曹正，高科科，薛维超 . 变频干衣机烘干性能及预测方法的理论与实验研究［J］. 制冷学报， 2025 ， 46 （ 2 ）： 155 - 161 .

Cao Zheng ， Gao Keke ， Xue Weichao . Theoretical and experimental study on drying performance and prediction model of heat pump clothes dryer ［J］. Journal of Refrigeration ， 2025 ， 46 （ 2 ）： 155 - 161 .

刘永利，邵双全 . 天然制冷剂R290用于热泵干衣机能效实验研究［J］. 制冷学报， 2022 ， 43 （ 4 ）： 69 - 76 .

Liu Yongli ， Shao Shuangquan . Experimental study on energy efficiency of heat pump clothes dryer with natural refrigerant R290 ［J］. Journal of Refrigeration ， 2022 ， 43 （ 4 ）： 69 - 76 .

段海洋，章松发，张阳，等 . 干衣机负载量分档识别的深度学习算法应用［C］// 2025年中国家用电器技术大会论文集， 2025 ： 11 - 16 .

Duan Haiyang ， Zhang Songfa ， Zhang Yang ， et al . Application of deep learning algorithms for load level classification and identification in clothes dryers ［C］// 2025 China Household Appliances Technology Conference ， 2025 ： 11 - 16 .

李鸿鹏，于彩侠 . 基于软件算法提升洗衣干衣机智能判干性能的尝试与研究［J］. 家电科技， 2022 （ S1 ）： 686 - 692 .

Li Hongpeng ， Yu Caixia . Attempt and research on improving intelligent drying performance of household washer and dryers based on software algorithm ［J］. Journal of Appliance Science & Technology ， 2022 （ S1 ）： 686 - 692 .

魏嘉玥，单泠璇，王忠卿，等 . 基于MEMS温湿度传感器的小负载智能判干研究［C］// 2025年中国家用电器技术大会论文集， 2025 ： 284 - 289 .

Wei Jiayue ， Shan Lingxuan ， Wang Zhongqing ， et al . Research on intelligent dryness detection for small-load using MEMS temperature and humidity sensors ［C］// 2025 China Household Appliances Technology Conference ， 2025 ： 284 - 289 .

王召兵，赵弇锋，徐雯婷，等 . 电容智能判干在家电的研究及应用［C］// 2024年中国家用电器技术大会论文集， 2024 ： 219 - 227 .

Wang Zhaobing ， Zhao Yanfeng ， Xu Wenting ， et al . Research and application of capacitor intelligent stem in home appliances ［C］// 2024 China Household Appliances Technology Conference ， 2024 ： 219 - 227 .

吕艳芬，许升，宋华诚，等 . 红外判干技术在干衣机上的研究与应用［C］// 2023年中国家用电器技术大会论文集， 2023 ： 1199 - 1205 .

Lv Yanfen ， Xu Sheng ， Song Huacheng ， et al . Research and application of infrared temperature measurement to judge the drying clothes ［C］// 2023 China Household Appliances Technology Conference ， 2023 ： 1199 - 1205 .

李日新，刘和成，欧汝浩，等 . 基于多层感知机的干衣机智能湿度预测模型研究［C］// 2021年中国家用电器技术大会论文集， 2021 ： 1104 - 1109 .

Li Rixin ， Liu Hecheng ， Ou Ruhao ， et al . Research on intelligent humidity forecasting model of clothes dryer based on multi-layer perceptron ［C］// 2021 China Household Appliances Technology Conference ， 2021 ： 1104 - 1109 .

Zambonin G ， Altinier F ， Beghi A ， et al . Machine learning-based soft sensors for the estimation of laundry moisture content in household dryer appliances ［J］. Energies ， 2019 ， 12 （ 20 ）： 3843 .

Hernandez-Bautista A J ， Cetina-Quiñones A J ， Messina S ， et al . Predictive modeling and sensitivity analysis to optimize the efficiency and operation conditions of the drying process ［J］. Journal of the Brazilian Society of Mechanical Sciences and Engineering ， 2025 ， 48 （ 1 ）： 13 .

Lv Qing ， Chen Shiqi ， Fu Chao ， et al . PO-BP neural network-based temperature prediction for tobacco drying machines ［J/OL］. Measurement and Control ， 2026 . https：//journals.sagepub.com/doi/10.1177/00202940251408648 https://journals.sagepub.com/doi/10.1177/00202940251408648 .

Dai Jianwu ， Du Yingqing ， Zeng Qi ， et al . Evaluation and modeling of drying kinetics and quality changes of bamboo shoots under air-impingement drying based on parameter intelligent control ［J］. LWT-Food Science and Technology ， 2026 ， 241 ： 119110 .

Madrigal S ， Gallinad R ， Vicario J L ， et al . Improving energy distribution in collective self-consumption via XGBoost-based allocation coefficients prediction ［J］. Applied Energy ， 2026 ， 409 ： 127469 .

Zhu Xinyi ， Shen Xiaoyan ， Chen Kailiang ， et al . Research on the prediction and influencing factors of heavy duty truck fuel consumption based on LightGBM ［J］. Energy ， 2024 ， 296 ： 131221 .

Ramalingam K . Ai-driven optimization and engine evaluation of Palmyra biodiesel synthesized via cobalt chromite catalysis ［J］. Energy Conversion and Management ， 2026 ， 352 ： 121100 .

赵嘉俊，单联瑜，陈壮，等 . R290热泵干衣机蒸发器流动传热特性研究［J］. 流体机械， 2026 ， 54 （ 1 ）： 84 - 92 .

Zhao Jiajun ， Shan Lianyu ， Chen Zhuang ， et al . Study on flow and heat transfer characteristics of R290 heat pump dryer evaporator ［J］. Fluid Machinery ， 2026 ， 54 （ 1 ）： 84 - 92 .

Mohammed R H ， Bhagwat R ， Schmid M ， et al . Development and lab testing of an efficient semi-open heat pump dryer ［C］// 2026 ASHRAE Winter Conference ， 2026 .

GB/T 20292—2019 家用滚筒式干衣机性能测试方法［S］. （GB/T 20292 —

2019 Methods for measuring the performance of tumble dryers for household use ［S］.

Nii-Armah Mettle H ， Osei G ， Winchester E A ， et al . Enhancing telecom customer churn prediction through hyperparameter tuning and soft-voting ensemble modeling ［J］. Applied Computational Intelligence and Soft Computing ， 2026 ， 2026 ： 7620084 .

舒展，马依兰，聂凯峰，等 . 基于OOD技术的网络告警日志高置信度研判方法［J］. 信息网络安全， 2026 ， 26 （ 4 ）： 626 - 641 .

Shu Zhan ， Ma Yilan ， Nie Kaifeng ， et al . A high-confidence assessment method for network alarm logs based on OOD technology ［J］. Netinfo Security ， 2026 ， 26 （ 4 ）： 626 - 641 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Theoretical and Experimental Study on Drying Performance and Prediction Model of Heat Pump Clothes Dryer

Experimental Study on Energy Efficiency of Heat Pump Clothes Dryer with Natural Refrigerant R290

Performance Comparison of Three Booster-Assisted Ejector Refrigeration Cycles with Environmentally Friendly Refrigerants

Enhancing HCFC-141b Hydrate Cold Thermal Energy Storage Density by EL Series Surfactants in Quiescent Conditions

Heat-Transfer Performance of Gas-Gap Heat Switch for mK Sorption Cooler

Related Author

Cao Zheng

Gao Keke

Xue Weichao

Liu Yongli

Related Institution

Midea Corporate Research Center

School of Energy and Power Engineering, Huazhong University of Science and Technology

Address：10/F, Yindu Mansion, 67 Fucheng Road, Haidian Dist, Beijing, China Postal code：100142
Tel：010-68711412 Email：editor@car.org.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰

AI伴读