solsticestudioai/solstice-residential-energy-pack

Solstice Residential Energy Pack (Sample)

A synthetic residential solar-plus-storage operations dataset for VPP dispatch, tariff-aware savings, billing, and outage resilience. This sample is designed for product demos, analytics workflows, dashboard prototyping, and AI model validation where real customer or utility data is unavailable or too sensitive to use.

Built by SolsticeAI as a free sample of a larger commercial pack. 100% synthetic. No real customer, meter, or utility records.

What is included

| File | Rows | Grain | Purpose |

|---|---:|---|---|

| households.csv | 500 | household | Household archetypes, geography, electrification, and outage risk |

| daily_generation_consumption.csv | 90,000 | date x household | Load, solar generation, import/export, battery usage, and daily savings |

| dispatch_events.csv | 5,819 | dispatch event | Requested vs delivered dispatch, participation, incentives, and grid value |

| billing_and_savings.csv | 3,000 | month x household | Counterfactual bills, subscription payments, credits, and net customer value |

| metric_definitions.csv | 3 | metric | Metric formulas and table-level documentation |

| dashboard_suggestions.csv | 3 | chart | Starter dashboard recipes for product and analytics teams |

Coverage: USA

Period: 6 months (2025-01-01 to 2025-06-29)

Join key: household_id

Formats in this sample repo: CSV

Why this dataset is useful

Most public solar or energy datasets are either too generic, too narrow, or detached from the operating model of a residential energy business. This sample is shaped around the questions a solar-plus-storage platform, VPP operator, DERMS vendor, or energy analytics team actually cares about:

• —Which household profiles create the highest dispatch value?

• —How much do tariff design and load shape affect savings?

• —Which homes deliver the most outage resilience value?

• —How reliable is dispatch participation across a residential fleet?

• —How do billing, credits, and contract economics affect customer value?

What makes the sample credible

• —Stable relational keys and business-readable tables

• —Daily operational energy facts rather than flat summary rows

• —Dispatch, billing, and savings data tied to the same household base

• —Structured for dashboarding, workflow testing, demos, and model development

• —Synthetic by design, so it can be shared safely across internal and external teams

Typical use cases

• —Residential energy product demos

• —VPP dispatch and participation analytics

• —Tariff-aware savings analysis

• —Billing workflow and customer-value testing

• —Outage resilience reporting

• —AI model validation on structured energy operations data

• —Dashboard and BI template development

Quick start

import pandas as pd





households = pd.read_csv("data/households/train.csv")


daily = pd.read_csv("data/daily_generation_consumption/train.csv", parse_dates=["date"])


dispatch = pd.read_csv("data/dispatch_events/train.csv", parse_dates=["date"])


billing = pd.read_csv("data/billing_and_savings/train.csv")





# Example: average savings by state


savings_by_state = (


    daily.merge(households[["household_id", "state"]], on="household_id", how="left")


         .groupby("state")["customer_savings_usd"]


         .mean()


         .reset_index()


)

Schema

See SCHEMA.md for the full field definitions and pack design.

See manifest.json for sample generation metadata and row counts.

License

Released under CC BY 4.0. Use freely for demos, internal tooling, research, education, and commercial prototyping with attribution.

Synthetic data only. No real customer, patient, meter, or utility information.

Get the full pack

This Hugging Face repo is a 500-household, 6-month sample. The production pack scales to 5,000–25,000+ households, 12+ month historical windows, additional tables (tariffs, outage events, service tickets, contracts, installations, enrollment, portfolio KPIs), CSV and Parquet delivery, and buyer-specific variants.

Self-serve (Stripe checkout):

• —**Sample Scale tier — $5,000** — ~25K records, one subject, 72-hour delivery.

Full pack + enterprise scope:

• —www.solsticestudio.ai/datasets — per-SKU pricing across Starter / Professional / Enterprise tiers, plus commercial licensing, custom generation, and buyer-specific variants.

Procurement catalog:

• —SolsticeAI Data Storefront — available via Datarade / Monda.

Citation

@dataset{solstice_residential_energy_pack_2026,


  title        = {Solstice Residential Energy Pack (Sample)},


  author       = {SolsticeAI},


  year         = {2026},


  publisher    = {Hugging Face},


  url          = {https://huggingface.co/datasets/solsticestudioai/solstice-residential-energy-pack}


}