Data-driven Prediction of Vehicle Air Conditioner Set Temperature
-
摘要: 为了对用户期望的车载空调温度进行实时预测, 本文提出了一种习惯温度预测模型和时间序列温度预测模型双模型耦合的方法对车载空调设定温度进行实时预测。该方法以车内和外界的多维度信息作为输入, 通过过滤式和随机森林对特征进行筛选, 并根据实际应用场景集成模型来对用户期望的空调设定温度进行预测。最后使用该模型对测试数据进行验证。结果表明本文提出的双模型耦合的方法对用户空调设定温度的预测结果平均绝对百分比误差(MAPE)为0.049, 能够精确地对车载空调温度进行预测, 从而为智能化、个性化调控空调提供辅助决策。
-
关键词:
- 习惯温度预测模型 /
- 时间序列温度预测模型 /
- 应用场景 /
- 集成模型 /
- 个性化
Abstract: In order to predict the real-time temperature of the vehicle air conditioner a user desires, this paper proposes a method that is coupled with the dual models of habit temperature prediction and time series temperature prediction to predict the real-time temperature of the vehicle air conditioner. The method takes the multi-dimensional information on the car and the information on the outside world as input, filters the features through filtering and random forest, and predicts the user's desired air-conditioning set temperature according to the actual application scenario integrated in the model. Finally, the model is used to verify the test data. The results show that the dual model-coupled method predicts the mean absolute percentage error (MAPE) of the user's vehicle air-conditioning set temperature to be as accurate as 0.049, thus providing auxiliary information on decision-making for intelligent and personalized air-conditioning control. -
表 1 数据名称说明
分类 特征字符 数据类型 说明 时间信息类 starttime int 数据采集时间 地理位置类 Latitude float 经度 Longitude float 纬度 空调参数类 vehac int 空调开关 vehacauto int 空调自动挡开关 vehaccirctype int 空调循环模式 vehacdrvtargettemp float 空调设定温度 环境信息类 vehoutsidetemp float 车外温度 vehinsidetemp float 车内温度 vehraindetected int 下雨检测 车辆信息类 VIN string 车辆识别号码 vehtype string 车型 vehsyspwrmod int 系统电源模式 表 2 数据预处理后的数据特征说明
分类 特征字符 数据类型 说明 时间类 year int 数据采集时间所属年 month int 数据采集时间所属月 hour int 数据采集时间所属一天中第几时 weekofyear int 数据采集时间所属一天中第几周 dayofweek int 数据采集时间所属一周中第几天 is-weekend int 数据采集时间是否是周末 is-night int 数据采集时间是否是夜晚 duration-time float 空调自每次开启的累计使用时长 地理位置类 geo-cluster int 四大地理分区 空调参数类 vehacdrvtargettemp float 空调设定温度 vehacdrvtargettemp-shift float 上一时刻的空调设定温度 环境类 vehoutsidetemp float 车外温度 vehinsidetemp float 车内温度 vehraindetected int 下雨检测 diff-temp-daynight float 昼夜温差 aver-temp float 平均日温 aver-wind int 平均日风速 weather-cluster int 天气状况 用户喜好类 favorite-type int 冷热喜好 人体感受类 temp-hm float 体感温度 表 3 性能评价指标
模型 评价指标MAPE RF 0.047 Lasso 0.022 RF-Lasso 0.049 -
[1] 工信部印发《工业控制系统信息安全行动计划(2018~2020年)》[J]. 信息技术与标准化, 2018(S1): 11The Ministry of Industry and Information Technology issued the 《Industrial Control System Information Security Action Plan (2018~2020)》[J]. Information Technology & Standardization, 2018(S1): 11 (in Chinese) [2] 宋洪健. 汽车空调控制系统优化及试验研究[D]. 杭州: 浙江大学, 2018SONG H J. Optimization and experimental study of automobile air conditioning control system[D]. Hangzhou: Zhejiang University, 2018 (in Chinese) [3] 张载龙, 茹亮. 基于BP神经网络的冷藏车温度预测研究[J]. 计算机技术与发展, 2013, 23(10): 180-183 https://www.cnki.com.cn/Article/CJFDTOTAL-WJFZ201310045.htmZHANG Z L, RU L. Study on prediction of temperature of refrigerated trucks based on BP neural network[J]. Computer Technology and Development, 2013, 23(10): 180-183 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-WJFZ201310045.htm [4] 翁建华, 舒宏坤, 石梦琦, 等. 热网络法预测车内温度的理论和实验研究[J]. 上海理工大学学报, 2018, 40(6): 552-556 https://www.cnki.com.cn/Article/CJFDTOTAL-HDGY201806007.htmWENG J H, SHU H K, SHI M Q, et al. Theoretical and experimental study on the thermal network method for temperature prediction in a car[J]. Journal of University of Shanghai for Science and Technology, 2018, 40(6): 552-556 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HDGY201806007.htm [5] 刘荣, 童亮, 许永红. 基于pso_FSVM的车用动力电池温度预测模型研究[J]. 现代电子技术, 2018, 41(12): 24-27 https://www.cnki.com.cn/Article/CJFDTOTAL-XDDJ201812007.htmLIU R, TONG L, XU Y H. Research on temperature prediction model of vehicle power battery based on pso_FSVM[J]. Modern Electronics Technique, 2018, 41(12): 24-27 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XDDJ201812007.htm [6] 殷青, 张岩, 韩昀松. 基于LSTM算法的严寒地区办公建筑过渡季室内温度预测模型构建[J]. 低温建筑技术, 2019, 41(3): 8-12 https://www.cnki.com.cn/Article/CJFDTOTAL-DRAW201903003.htmYIN Q, ZHANG Y, HAN J S. An indoor temperature prediction model for office building in transition season in severe cold region based on LSTM[J]. Low Temperature Architecture Technology, 2019, 41(3): 8-12 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DRAW201903003.htm [7] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016ZHOU Z H. Machine learning[M]. Beijing: Tsinghua University Press, 2016 (in Chinese) [8] SCHIAVON S, HOYT T, PICCIOLI A. Web application for thermal comfort visualization and calculation according to ASHRAE standard 55[J]. Building Simulation, 2014, 7(4): 321-334 doi: 10.1007/s12273-013-0162-3 [9] 楼海军, 阚安康, 康利云, 等. 船舶舱室空调热舒适性评价指标及其微气候参数优化[J]. 船舶工程, 2014, 36(S1): 80-83, 90 https://www.cnki.com.cn/Article/CJFDTOTAL-CANB2014S1023.htmLOU H J, KAN A K, KANG L Y, et al. Thermal comfort index of marine cabin air-conditioning system and the optimization of cabin climate[J]. Ship Engineering, 2014, 36(S1): 80-83, 90 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-CANB2014S1023.htm [10] INGERSOLL J G, KALMAN T G, MAXWELL L M, et al. Automobile passenger compartment thermal comfort model-Part Ⅱ human thermal comfort calculation[M]. Detroit, Michigan, United States: International Congress & Exposition, 1992 [11] FANGER P O. Thermal comfort: analysis and applications in environmental engineering[M]. New York: McGraw-Hill Book Company, 1973 [12] FANGER P O, MELIKOV A K, HANZAWA H, et al. Air turbulence and sensation of draught[J]. Energy and Buildings, 1988, 12(1): 21-39 doi: 10.1016/0378-7788(88)90053-9 [13] 韩滔. 基于动态热舒适的空调控制方案研究[D]. 成都: 西南交通大学, 2006HAN T. Study on dynamic thermal-based air-conditioning control scheme[D]. Chengdu: Southwest Jiaotong University, 2006 (in Chinese) [14] WONG T T. Performance evaluation of classification algorithms by k-fold and leave-one-out cross validation[J]. Pattern Recognition, 2015, 48(9): 2839-2846 doi: 10.1016/j.patcog.2015.03.009 [15] 邓帅. 基于改进贝叶斯优化算法的CNN超参数优化方法[J]. 计算机应用研究, 2019, 36(7): 1984-1987 https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ201907014.htmDENG S. Hyper-parameter optimization of CNN based on improved Bayesian optimization algorithm[J]. Application Research of Computers, 2019, 36(7): 1984-1987 (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSYJ201907014.htm [16] 朱明敏. 贝叶斯网络结构学习与推理研究[D]. 西安: 西安电子科技大学, 2013ZHU M M. Research on structural learning and inference in Bayesian networks[D]. Xi'an: Xidian University, 2013 (in Chinese)