ISONE Smart City Energy Dataset - Real Urban Electricity System Data from ISO New England

Overview

The ISONE Smart City Energy Dataset published on Kaggle in May 2025 provides authentic operational data from the Independent System Operator of New England (ISO-NE), offering insights into real-world urban electricity systems.

Data Source and Coverage

• Derived from ISO New England's publicly available operational data, covering electricity generation, consumption, and market activity across the New England region (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont).
• Focus on urban and suburban load zones representing major metropolitan areas (Boston, Hartford, Providence) and their electricity demand patterns.
• Temporal coverage spanning multiple years with granularity ranging from 5-minute real-time data to hourly and daily aggregates.

Key Variables

• Electricity demand (load): Total and zonal electricity consumption in megawatts (MW), broken down by load zone and time.
• Generation mix: Power generation by source (natural gas, nuclear, hydro, wind, solar, oil, coal) showing the renewable energy penetration and fossil fuel dependency.
• Locational Marginal Prices (LMP): Real-time and day-ahead electricity prices at different nodes, reflecting supply-demand balance and transmission constraints.
• Grid operations: Frequency regulation, spinning reserves, and ancillary services data indicating grid stability and reliability.
• Weather correlations: Temperature, humidity, and other meteorological data matched to load zones for weather-sensitive demand modeling.

Smart City Context

• Data enables analysis of urban energy consumption patterns, peak demand drivers (cooling in summer, heating in winter), and the impact of distributed energy resources (rooftop solar, electric vehicles).
• Supports research on demand response, time-of-use pricing effectiveness, and consumer behavior in response to real-time pricing signals.
• Provides baseline for evaluating smart grid investments and their impact on grid efficiency and renewable integration.

Use Cases

• Energy demand forecasting: Training models to predict short-term and long-term electricity demand for urban areas using historical load and weather data.
• Renewable energy integration: Analyzing variability and intermittency of wind and solar generation and their impact on grid operations.
• Price prediction: Developing algorithms to forecast electricity prices for optimizing energy procurement and consumption in smart buildings.
• Grid optimization: Identifying congestion patterns, transmission bottlenecks, and opportunities for infrastructure upgrades.
• Policy analysis: Evaluating the effectiveness of renewable energy mandates, carbon pricing, and demand response programs on urban energy systems.