A Real-World Energy Management Dataset from a Smart Company Building for Optimization and Machine Learning
🤖 AI Summary
To address the dual challenges of intelligent building operations optimization and carbon reduction in enterprise campuses, this study constructs and open-sources the first industrial-grade, six-year (2018–2023) energy management dataset—comprising 72 electricity meters, 9 heat meters, and on-site weather station data. The dataset features a hierarchical metering architecture, end-to-end coverage (including photovoltaic/CHP generation, energy consumption, environmental control, and meteorology), and expert-annotated anomalous events, supporting multi-granularity time-series processing (raw/processed/labeled). Integrating physics-informed modeling, optimization-based scheduling, and deep learning, our approach achieves significant improvements in real-world deployment: short-term load forecasting MAE reduced by 12.3%, chiller/boiler scheduling energy consumption decreased by 8.7%, and carbon emission estimation accuracy enhanced (R² increased by 0.15). Our core contribution is the first long-term, heterogeneous, and comprehensively annotated building energy benchmark dataset—enabling reproducible, verifiable AI research for low-carbon buildings.
Application Category
📝 Abstract
We present a large real-world dataset obtained from monitoring a smart company facility over the course of six years, from 2018 to 2023. The dataset includes energy consumption data from various facility areas and components, energy production data from a photovoltaic system and a combined heat and power plant, operational data from heating and cooling systems, and weather data from an on-site weather station. The measurement sensors installed throughout the facility are organized in a hierarchical metering structure with multiple sub-metering levels, which is reflected in the dataset. The dataset contains measurement data from 72 energy meters, 9 heat meters and a weather station. Both raw and processed data at different processing levels, including labeled issues, is available. In this paper, we describe the data acquisition and post-processing employed to create the dataset. The dataset enables the application of a wide range of methods in the domain of energy management, including optimization, modeling, and machine learning to optimize building operations and reduce costs and carbon emissions.
Problem
Research questions and friction points this paper is trying to address.
Provides real-world energy data for smart building optimization.
Enables machine learning to reduce energy costs and emissions.
Supports modeling of energy systems using hierarchical metering data.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Hierarchical metering structure for energy data
Integration of photovoltaic and heat power data
Multi-level processed data for machine learning
🔎 Similar Papers
