汽车工程 ›› 2024, Vol. 46 ›› Issue (6): 1114-1124.doi: 10.19562/j.chinasae.qcgc.2024.06.018

• • 上一篇    下一篇

汽车乘员舱个性化热舒适智能空调决策系统设计

罗杨1,刘平1(),刘明杰1,朴昌浩1,袁朋2,万凯林2   

  1. 1.重庆邮电大学自动化学院,重庆 400065
    2.重庆长安汽车股份有限公司,重庆 400023
  • 收稿日期:2023-09-18 修回日期:2023-12-26 出版日期:2024-06-25 发布日期:2024-06-19
  • 通讯作者: 刘平 E-mail:liuping_cqupt@cqupt.edu.cn
  • 基金资助:
    国家重点研发计划项目(2022YFE0101000);重庆市自然科学基金面上项目(CTSB2022NSCQ-MSX0355)

Decision System Design of Personalized Thermal Comfortable Intelligent Air Conditioning for Automobile Passenger Cabin

Yang Luo1,Ping Liu1(),Mingjie Liu1,Changhao Piao1,Peng Yuan2,Kailin Wan2   

  1. 1.College of Automation,Chongqing University of Posts and Telecommunications,Chongqing  400065
    2.Chongqing Changan Automobile Co. ,Ltd. ,Chongqing  400023
  • Received:2023-09-18 Revised:2023-12-26 Online:2024-06-25 Published:2024-06-19
  • Contact: Ping Liu E-mail:liuping_cqupt@cqupt.edu.cn

摘要:

为了进一步提高汽车乘员舱空调系统的智能化和舒适性水平,本文提出了一种基于热舒适理论的个性化智能空调决策系统设计方案。首先,针对汽车乘员舱改进了基于PMV(predicted mean vote)和PPD(predicted percentage of dissatisfaction)理论的热舒适性计算方法;进一步,利用人体画像技术实现了乘员舱驾乘人员的热舒适性特征提取,并在专家经验知识的基础上构建了具有理论计算依据的乘员舱热舒适数据集;然后,利用机器学习算法搭建了个性化热舒适空调系统随机森林决策模型,以此满足个性化热舒适智能决策需求;最后,给出了完整的系统框架和设计。测试结果显示所提出的系统模型决策准确率在90%以上,实车测试结果表明:本文系统能够识别驾乘人员特征,实时进行个性化热舒适性参数推荐,验证了本研究决策方法的有效性和实用价值。

关键词: 汽车, 乘员舱, 个性化热舒适, 空调系统, 智能决策

Abstract:

In order to further improve the intelligence and comfort level of the air conditioning system of the automobile passenger cabins, a personalized intelligent air conditioning decision system design based on the thermal comfort theory is proposed in this paper. Firstly, the thermal comfort calculation method based on the predicted mean vote (PMV) and predicted percentage of dissatisfaction (PPD) theories is improved for the automobile passenger cabin. Furthermore, the thermal comfort features of passengers in the passenger cabin are extracted by using human portrait technology, and the theoretical calculation-based thermal comfort data set of the passenger cabin is constructed on the basis of experts' experience knowledge. Then, machine-learning algorithm is used to build the random forest decision-making model of personalized thermal comfort air conditioning system, so as to meet the intelligent decision-making requirements of personalized thermal comfort. Finally, the complete system framework and design are given. The test results show that the decision-making accuracy of the proposed system model is above 90%, and the results of real vehicle testing show that the proposed system can identify the characteristics of drivers and passengers to recommend personalized thermal comfort parameters in real time, which verifies the effectiveness and practical value of the decision-making method in this study.

Key words: automobile, passenger cabin, personalized thermal comfortable, air conditioning system, intelligent decision-making