大数据驱动的城市居民低碳行为画像与碳普惠激励机制研究OA
Research on low-carbon behavior profiling and carbon-inclusive incentive mechanisms for urban residents driven by big data
文章以国内四线城市32.6万名居民2024年全年多源行为数据为基础,融合能源消耗、出行轨迹、绿色消费与社区参与4类17项指标,运用K-Means聚类算法构建低碳行为画像体系,将居民划分为全域低碳高践行、出行低碳主导、家庭节能主导、潜力可转化及高碳惯性5类典型群体.研究发现,各群体在行为短板、减排潜力与行为稳定性上呈现显著分化,其中潜力可转化群体减排空间最大(人均月减排10.36 kg),高碳惯性群体占比最高(26.68%)且稳定性最低.据此设计差异化碳积分核算规则与分级激励方案,为城市居民端低碳治理提供量化分析工具和政策设计参考.
Based on multi-source behavioral data from 326 000 residents in a fourth-tier Chinese city throughout 2024,this study integrates four categories and seventeen indicators covering energy consumption,travel trajectories,green consumption,and community participation.The K-Means clustering algorithm is employed to construct a low-carbon behavior profiling system,which classifies residents into five typical groups:full-domain high practice,travel-led low carbon,household energy saving dominant,potential convertible,and high-carbon inertial.The findings reveal significant differentiation among these groups in terms of behavioral shortcomings,emission reduction potential,and behavioral stability.Among them,the potential convertible group exhibits the largest emission reduction space(10.36 kg per capita per month),while the high-carbon inertial group accounts for the highest proportion(26.68%)and shows the lowest stability.Accordingly,differentiated carbon credit accounting rules and tiered incentive schemes are designed,providing quantitative analytical tools and policy design references for low-carbon governance at the urban resident level.
王鹏;皇甫政彤;许峰
海南科技职业大学,海南 海口 570100||东北农业大学经济管理学院,黑龙江 哈尔滨 150030海南科技职业大学,海南 海口 570100||领熵科学技术研究院科技成果处,福建 厦门 361015海南科技职业大学,海南 海口 570100
资源环境
城市居民低碳行为行为画像碳普惠激励机制
urban residentslow-carbon behaviorbehavior profilingcarbon inclusionincentive mechanism
《智能城市》 2026 (8)
1-6,6
2022年海南省哲学社会科学规划课题(HNSK(YB)22-23)2025年中国城市商业网点建设管理联合会课题(CUCO2025LXYK0041)
评论