Integrated Dataset of Train Energy Consumption and Meteorological Dataset

This dataset supports the study “Probabilistic Profiling of Train-Generated Regenerative-Braking Electricity under Partial Observability: A Semiparametric Mixed-Effects Factor Model.” It contains merged train-operation, regenerative-braking, meteorological, and passenger-flow data for 15 trainsets (JT001–JT015) operating on the Chongqing Jiangtiao through-service corridor. The workbook is organized into 15 worksheets, one for each trainset, and includes 2,205 train-day records covering 2 January to 30 November 2025. Each record contains the operating date, daily train-generated regenerative-braking electricity, daily mileage, daily mean temperature, daily mean relative humidity, daily precipitation, and aggregate daily passenger flow. The train-operation and regenerative-braking records were provided by the collaborating enterprise and were cleaned, organized, and merged for research use. Meteorological variables were obtained from the ERA5-Land reanalysis dataset and spatially matched to the corridor. Passenger-flow data represent aggregate corridor-level demand and are used as a proxy for passenger-related operating intensity rather than direct train-specific passenger loading. The corridor spans both 25 kV AC and 1500 V DC traction-supply environments; however, the available enterprise records report only full-corridor daily regenerative-braking totals, so section-specific contributions cannot be separated. In the associated study, daily regenerative-braking electricity is normalized by daily mileage to reduce first-order differences in operating exposure. Records with daily mileage between 100 and 1,000 km are retained as the regular-service analysis sample, yielding 2,021 valid train-day observations. The dataset is intended to support reproducibility of the train-level probabilistic profiling, heterogeneity assessment, and model-based counterfactual analyses. Regenerative-braking electricity refers to electricity generated at the train side and should not be interpreted as system-level recovered energy or grid electricity savings.本数据集用于支撑题为《部分可观测条件下列车再生制动发电量的概率画像：半参数混合效应因子模型》的研究。本数据集包含运行于重庆江跳直通运营线路的15列列车组（JT001–JT015）的列车运行、再生制动发电量（regenerative-braking electricity）、气象及客流融合数据。该工作簿包含15个工作表，每列列车组对应一个工作表，共收录2205条列车日度记录，时间范围为2025年1月2日至2025年11月30日。 每条记录包含运行日期、当日列车再生制动发电量、当日运营里程、当日平均气温、当日平均相对湿度、当日降水量及当日总客流。列车运行及再生制动相关数据由合作企业提供，并经过清洗、整理与融合后供研究使用。气象变量取自ERA5-Land再分析数据集，并通过空间匹配关联至本线路。客流数据为线路层面的总客流需求，被用作与旅客相关的运营强度的代理变量，而非单列车的具体载客量。 本线路同时覆盖25 kV交流与1500 V直流牵引供电场景，但现有企业数据仅提供线路层面的日度再生制动总发电量，无法拆分各区间的贡献值。在本关联研究中，为降低运营暴露度的一阶差异，对日度再生制动发电量以当日运营里程进行归一化处理。本研究将当日运营里程介于100至1000 km之间的记录作为常规运营分析样本，最终得到2021条有效列车日度观测数据。 本数据集旨在支撑列车层面概率画像、异质性评估及基于模型的反事实分析的可复现性研究。 注：再生制动发电量指列车侧产生的电能，不应被解读为系统层面的回收能源或电网节电收益。