机器学习算法模型为住宅低碳设计与优化提供了数据支持。然而, 在碳排放预测与分析时, 算法模型常被直接使用, 而未考虑调参与寻优, 且不同自变量数据集对模型预测效果的影响差异也有待明确。为揭示不同算法模型对寒冷地区住宅低碳设计的指导效果、 向建筑师提供算法模型的选择依据, 针对多元线性回归、 分类回归树、 随机森林、 自适应增强算法、 梯度提升回归树和多层感知机等在低碳设计中常用的算法模型进行寻优, 对比分析不同算法和自变量数据集的适用性与预测性能。该文说明了算法模型寻优的目标边界、 参数取值范围、 寻优过程和论证方法。在37栋寒冷地区钢筋混凝土剪力墙结构住宅及其衍生方案的基础上, 采用交叉验证和网格搜索, 建立了120个建材碳排放预测模型和60个将稳态耗热量转换为动态耗热量的转化系数预测模型。对比结果表明： 总体上, 多元线性回归、 随机森林和梯度提升回归树算法的碳排放预测性能更好。其中, 随机森林和梯度提升回归树算法在误差控制方面表现更佳, 但预测优度与多元线性回归算法相近, 且可解释性较差。采用恰当的自变量数据集, 如建筑总层数、 建筑层高、 建筑面宽与进深等形体尺度参数, 标准层户数与卧室数等功能配置参数, 以及采暖期室外平均温度、 实际供暖天数、 屋面和墙面传热系数的修正系数等城市气象参数, 多元线性回归算法能够为寒冷地区住宅低碳设计与优化提供更直观、 有效的指导。
Abstract
[Objective] Machine learning algorithms provide valuable data support for designing and optimizing low-carbon residential buildings. However, when used directly for carbon emission prediction and analysis, these models often lack proper parameter tuning and optimization. The different impacts of various independent variable datasets on predictive performance also remain to be clarified. In China's cold zones, where residential buildings share similar architectural structures, energy-saving designs, and spatial layouts, carbon emissions primarily come from the operational phase and the production stages of building materials, with heating emissions being a significant component. This study aims to elucidate the effectiveness of different machine learning algorithm models in guiding low-carbon residential design in these cold zones, offering architects criteria for selecting proper algorithms. This study focuses on automatic parameter tuning and optimization for several commonly used algorithms in the context of low-carbon design of buildings, including multiple linear regression, classification and regression tree, random forest, adaptive boosting, gradient boosting regression tree, and multilayer perceptron. The study compares and analyzes the performance limits and applicability of these algorithms and independent variable datasets in predicting carbon emissions during building material production and heating stages. [Methods] This paper elaborates on the target boundaries, parameter ranges, optimization processes, and validation methods for optimizing machine learning algorithm models. Through comprehensive research and simulation analysis of 37 reinforced concrete shear wall residential buildings and their derivative schemes in cold zones, multiple independent variable datasets suitable for establishing predictive models are identified. Cross-validation and grid search techniques are employed to optimize the predictive performance limits of different machine learning algorithms and independent variable datasets. Subsequently, 120 models for predicting carbon emissions from building materials and 60 models for transforming steady-state heating consumption into dynamic heating consumption using the six mentioned algorithms are established. [Results] A horizontal comparison of the models reveals that algorithms such as multiple linear regression, random forest, and gradient boosting regression trees exhibit relatively good performance (R2 over 0.900) in carbon emission prediction after hyperparameter tuning across different independent variable datasets. Random forest and gradient boosting regression tree models excel in error control and offer similar predictive accuracy to multiple linear regression but lack interpretability. In contrast, multiple linear regression models provide clearer equations and stronger guidance for low-carbon design and optimization, focusing on carbon emission reduction during building material production or winter heating stages. Models based on the total residential building area exhibit optimal performance in predicting building material carbon emissions. Predictive models built on parameters such as the number of above-ground and underground floors, building width and depth, total household numbers, number of bedrooms for standard floor, and total number of residential bathrooms in the residence also demonstrate strong predictive capabilities for building material carbon emissions. For predicting the conversion coefficient during the heating stage, including the number of households and bedrooms per standard floor as independent variables significantly enhances predictive performance. [Conclusions] Although various machine learning models are useful for predicting residential building carbon emissions, the multiple linear regression model stands out owing to its excellent predictive performance and its intuitive representation of how design parameters affect carbon emissions. By utilizing different and appropriate independent variable datasets, such as the total number of floors, floor height, building dimensions, number of households and bedrooms on a floor, and corrected coefficients for urban meteorological parameters (including outdoor average temperature during the heating season, actual heating days, and roof and wall heat transfer coefficients), or by adopting the finally determined total building area, the multiple linear regression algorithm can deliver timely and multi-faceted guidance. These results are crucial for low-carbon design and optimization during the primary stages of the residential lifecycle in China's cold zones.
关键词
碳排放 /
机器学习 /
设计参数 /
交叉验证 /
网格搜索
Key words
carbon emission /
machine learning /
design parameters /
cross-validation /
grid search
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 刘依明, 刘念雄. 建筑生命周期评估中碳排放计算的重要意义[J]. 住区, 2020(3): 46-51. LIU Y M, LIU N X. The importance of carbon emission calculation in building life cycle assessment [J]. Design Community, 2020(3): 46-51. (in Chinese)
[2] 中国建筑节能协会, 重庆大学城乡建设与发展研究院. 中国建筑能耗与碳排放研究报告(2023年) [J]. 建筑, 2024(2): 46-59. China Association of Building Energy Efficiency; Institute of Urban-Rural Construction and Development, Chongqing University. Research report on China building energy consumption and carbon emissions (2023) [J]. Construction and Architecture, 2024(2): 46-59. (in Chinese)
[3] 李金潞. 寒冷地区城市住宅全生命周期碳排放测算及减碳策略研究[D]. 西安: 西安建筑科技大学, 2019. LI J L. Study on carbon emissions calculation and carbon reduction strategy of urban residential life cycle in cold areas [D]. Xi'an: Xi'an University of Architecture and Technology, 2019. (in Chinese)
[4] 田绅, 刘稷轩, 唐明生. 既有住宅建筑舒适性与节能技术研究[J]. 住宅科技, 2017, 37(1): 27-31. TIAN S, LIU J X, TANG M S. Research on the comfort and energy-saving technologies of existing residential buildings [J]. Housing Science, 2017, 37(1): 27-31. (in Chinese)
[5] MITCHELL T M. 机器学习[M]. 曾华军, 张银奎, 译. 北京: 机械工业出版社, 2003. MITCHELL T M. Machine learning [M]. ZHENG H J, ZHANG Y K, trans. Beijing: China Machine Press, 2003. (in Chinese)
[6] 袁梅宇. 机器学习基础: 原理、 算法与实践[M]. 北京: 清华大学出版社, 2018. YUAN M Y. Machine learning foundations: Principles, algorithms and practice [M]. Beijing: Tsinghua University Press, 2018. (in Chinese)
[7] 黄莉婷, 苏川集. 白话机器学习算法[M]. 武传海, 译. 北京: 人民邮电出版社, 2019. HUANG L T, SU C J. Numsense! Data science for the layman: No math added [M]. WU C H, trans. Beijing: Posts & Telecom Press, 2019. (in Chinese)
[8] 肖建, 于龙, 白裔峰. 支持向量回归中核函数和超参数选择方法综述[J]. 西南交通大学学报, 2008, 43(3): 297-303. XIAO J, YU L, BAI Y F. Survey of the selection of kernels and hyper-parameters in support vector regression [J]. Journal of Southwest Jiaotong University, 2008, 43(3): 297-303. (in Chinese)
[9] 吴涛, 贺汉根, 贺明科. 基于插值的核函数构造[J]. 计算机学报, 2003, 26(8): 990-996. WU T, HE H G, HE M K. Interpolation based kernel function's construction [J]. Chinese Journal of Computers, 2003(8): 990-996. (in Chinese)
[10] GOLD C, SOLLICH P. Model selection for support vector machine classification [J]. Neurocomputing, 2003, 55(1-2): 221-249.
[11] LAUER F, BLOCH G. Incorporating prior knowledge in support vector regression [J]. Machine Learning, 2008, 70(1): 89-118.
[12] BREIMAN L, FRIEDMAN J, OLSHEN R, et al. Classification and regression trees [M]. Belmont, USA: Wadsworth, 1984: 1-357.
[13] FRIEDMAN J H. Greedy function approximation: A gradient boosting machine [J]. The Annals of Statistics, 2001, 29(5): 1189-1232.
[14] SCHONLAU M. Boosted regression (boosting): An introductory tutorial and a Stata plugin [J]. The Stata Journal: Promoting Communications on Statistics and Stata, 2005, 5(3): 330-354.
[15] NATEKIN A, KNOLL A. Gradient boosting machines, a tutorial [J]. Frontiers in Neurorobotics, 2013, 7: 21.
[16] HUANG Y F, LIU Y H, LI C H, et al. GBRTVis: Online analysis of gradient boosting regression tree [J]. Journal of Visualization, 2019, 22(1): 125-140.
[17] 胡振, 龚薛, 刘华. 基于BP模型的西部城市家庭消费碳排放预测研究: 以西安市为例[J]. 干旱区资源与环境, 2020, 34(7): 82-89. HU Z, GONG X, LIU H. Prediction of household consumption carbon emission in western cities based on BP model: Case of Xi'an city [J]. Journal of Arid Land Resources and Environment, 2020, 34(7): 82-89. (in Chinese)
[18] 李锐, 刘鹏飞, 王岩. 基于决策树方法的建筑空调能耗数据分析[J]. 建筑节能, 2019, 47(8): 14-18. LI R, LIU P F, WANG Y. Data analysis of building air-conditioning energy consumption based on decision tree method [J]. Building Energy Efficiency, 2019, 47(8): 14-18. (in Chinese)
[19] 毛希凯. 建筑生命周期碳排放预测模型研究： 以天津市住宅为例[D]. 天津: 天津大学, 2018. MAO X K. Study on building life cycle carbon emissions prediction model: Taking residential buildings in Tianjin as examples [D]. Tianjin: Tianjin University, 2018. (in Chinese)
[20] 宋志茜. 建筑物化阶段碳排放特征及减碳策略研究[D]. 杭州: 浙江大学, 2023. SONG Z Q. Research on carbon emission characteristics and carbon reduction strategies in the materialization stage of architecture [D]. Hangzhou: Zhejiang University, 2023. (in Chinese)
[21] 刘佳丽. 直辖市下城镇住宅建筑二氧化碳排放量预测模型的应用研究[D]. 郑州: 河南工业大学, 2023. LIU J L. Application research on carbon dioxide emission prediction model of urban residential buildings in municipalities [D]. Zhengzhou: Henan University of Technology, 2023. (in Chinese)
[22] 黄耀. 基于集成学习的居民建筑能耗预测及模型优化[D]. 武汉: 华中科技大学, 2019. HUANG Y. Energy consumption prediction and model optimization of residential buildings based on ensemble learning [D]. Wuhan: Huazhong University of Science and Technology, 2019. (in Chinese)
[23] 步婷, 范蕊, 孙可欣, 等. 基于机器学习算法的区域建筑负荷预测建模研究[J]. 建筑科学, 2022, 38(4): 85-96. BU T, FAN R, SUN K X, et al. Research on modeling of regional cooling and heating load forecast based on machine learning algorithm [J]. Building Science, 2022, 38(4): 85-96. (in Chinese)
[24] 焦良珍, 陈海生, 高革, 等. 基于数据挖掘算法的DHC系统负荷时序预测方法[J]. 建筑节能, 2020, 48(11): 38-44. JIAO L Z, CHEN H S, GAO G, et al. Load estimation for the DHC system based on data mining and time-series techniques [J]. Building Energy Efficiency, 2020, 48(11): 38-44. (in Chinese)
[25] 程亚豪, 陈焕新, 王江宇. 基于机器学习的住宅能耗预测[J]. 制冷与空调, 2019, 19(5): 35-40. CHENG Y H, CHEN H X, WANG J Y. Prediction of residential energy consumption based on machine learning [J]. Refrigeration and Air-Conditioning, 2019, 19(5): 35-40. (in Chinese)
[26] 王志强, 任金哥, 韩硕, 等. 基于可解释性机器学习的建筑物物化阶段碳排放量预测研究[P]. 安全与环境学报, 2023： 1-13. (2023-11-03) [P]. https://doi.org/10.13637/j.issn.1009-6094.2023.1467. WANG Z Q, REN J G, HAN S, et al. Interpretable machine learning-based carbon emission prediction in the materialization stage of buildings [P]. Journal of Safety and Environment, 2023： 1-13. (2023-11-03). https://doi.org/10.13637/j.issn.1009-6094.2023.1467. (in Chinese)
[27] 王一帆, 李雪, 袁大昌. 住宅形态与碳排放相关性研究: 以金堂县为例[J]. 建筑节能, 2021, 49(9): 155-160. WANG Y F, LI X, YUAN D C. Influence of residential form on carbon emission: A case study on Jintang county [J]. Building Energy Efficiency, 2021, 49(9): 155-160. (in Chinese)
[28] 夏绪勇, 李书阳, 张永炜, 等. 建筑碳排放设计指南: 探求建筑行业碳中和路径[M]. 北京: 中国建筑工业出版社, 2023. XIA X Y, LI S Y, ZHANG Y W, et al. Building carbon emission design guidelines: Seeking a path to carbon neutrality in the construction industry [M]. Beijing: China Architecture & Building Press, 2023. (in Chinese)
[29] 刘依明, 刘念雄, 许沛琪. 寒冷地区住宅建材生产阶段碳排放预测模型[J]. 清华大学学报(自然科学版), 2023, 63(1): 15-23. LIU Y M, LIU N X, XU P Q. Carbon emission prediction model during the material production stage for cold zone residential buildings [J]. Journal of Tsinghua University (Science and Technology), 2023, 63(1): 15-23. (in Chinese)
[30] 中华人民共和国住房和城乡建设部. 建筑节能与可再生能源利用通用规范: GB 55015—2021[S]. 北京: 中国建筑工业出版社, 2022. Ministry of Housing and Urban-Rural Development of the People's Republic of China. General code for energy efficiency and renewable energy application in buildings: GB 55015—2021[S]. Beijing: China Architecture & Building Press, 2022. (in Chinese)
[31] 何泉, 盛昂昂, 刘大龙. 围护结构保温设计中非稳态计算方法适用性研究[J]. 西安建筑科技大学学报(自然科学版), 2021, 53(4): 561-567. HE Q, SHENG A A, LIU D L. Study on the applicability of unsteady calculation method in thermal insulation design of enclosure structure in cold area [J]. Journal of Xi'an University of Architecture and Technology (Natural Science Edition), 2021, 53(4): 561-567. (in Chinese)
[32] 中华人民共和国住房和城乡建设部. 严寒和寒冷地区居住建筑节能设计标准: JGJ 26—2018[S]. 北京: 中国建筑工业出版社, 2019. Ministry of Housing and Urban-Rural Development of the People's Republic of China. Design standard for energy efficiency of residential buildings in severe cold and cold zones: JGJ 26—2018[S]. Beijing: China Architecture & Building Press, 2019. (in Chinese)
[33] 中华人民共和国住房和城乡建设部. 建筑碳排放计算标准: GB/T 51366—2019[S]. 北京: 中国建筑工业出版社, 2019. Ministry of Housing and Urban-Rural Development of the People's Republic of China. Standard for building carbon emission calculation: GB/T 51366—2019[S]. Beijing: China Architecture & Building Press, 2019. (in Chinese)
[34] BAO Y K, LIU Z T. A fast grid search method in support vector regression forecasting time series [C]//Proceedings of the 7th International Conference on Intelligent Data Engineering and Automated Learning (IDEAL 2006). Burgos, Spain: Springer, 2006: 504-511.
基金
浙江省科技计划项目(2023C03173);国家重点研发计划项目(2022YFC3803800);国家自然科学基金重点项目(52130803)
PDF(6005 KB)
