Integrated Dataset of Train Energy Consumption and Meteorological Dataset

Published: 1 September 2026| Version 1 | DOI: 10.17632/d8nssm6kt8.1
Contributor:
JIE LIU

Description

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.

Files

Categories

Bayesian Statistics, Mathematical Optimization, Transportation Engineering

Licence