Norwegian Energy Community Dataset for Local Electricity-Hydrogen System Modeling and Optimization

1. System Description

The dataset represents a local energy community located in Porsgrunn, Norway, designed to explore integrated electricity and hydrogen systems under realistic operating conditions. The modeled community includes 400 electricity users across residential, commercial and industrial sectors, and are selected from the dataset created earlier. This mix reflects the diversity of demand profiles typically found in urban and semi-urban Norwegian locations.

Of the 400 users, 300 are assumed to be prosumers equipped with rooftop solar photovoltaic (PV) systems of varying capacities. These users contribute to local renewable electricity generation. Additionally, 250 users are equipped with behind-the-meter battery energy storage systems, adding flexibility and allowing for peak shaving or load shifting strategies. The battery energy storage system dataset is chosen from an original European dataset.

The system also includes a community-scale wind power generation unit, based on a Nordex N60 turbine with a rated capacity of 1.3 MW. This turbine location is set within the Porsgrunn region, and hourly wind power generation data is included in the dataset. The hourly power generation data is generated using the online simulation tool renewables.ninja.

Electricity consumption data for all 400 users is provided at an hourly resolution. The data originates from actual smart meter readings and captures real load patterns. Many residential users show characteristics consistent with electric vehicle (EV) ownership, adding variability to household demand profiles.

Electricity market interactions are modeled using hourly buy and sell prices for the year 2024. Buy prices are based on actual values provided in [5], while sell prices are derived from official statistics published by Statistics Norway, with values corresponding to the third quarter of 2024, excluding taxes and grid fees.

In parallel to the electricity system, the dataset models hydrogen demand and production. Hydrogen is primarily used for transportation purposes. The demand side includes two categories: A hydrogen-powered ferry fleet (6 ferries) based on a previous Norwegian study. A synthetic fleet of 15 hydrogen buses assigned to urban routes with varying lengths and frequencies.

Hydrogen is produced locally using two electrolyzers: a 1.2 MW system inspired by the HyBalance project in Denmark and a 2.5 MW system modeled after the Haeolus project in Norway. These electrolyzers are connected to the local electricity system and paired with hydrogen storage facilities, enabling time-shifted production and demand balancing.

2. Dataset Description

The dataset spans a 9-month period from January 1 to September 30, 2024 and includes hourly data across several categories:

Table 1. Dataset description

Dataset Component

Description

Gen_pv

Hourly PV generation data for 300 users. Generated by scaling 25 base profiles using random multipliers (0.8–1.2) to simulate diverse capacities.

Gen_wind

Hourly wind generation for a 1.3 MW turbine using data from Renewables.ninja (Nordex N60 in Porsgrunn).

Load

Hourly electricity consumption from 400 users (real smart meter data including EV-driven peaks).

Buy_price

Hourly electricity purchase price for 2024

Sell_price

Derived electricity selling price based on Q3 2024 values from Statistics Norway (excluding taxes/fees).

Bat

Battery capacities and performance parameters for 250 users using 8 known battery models

H2_ferry_demand

Synthetic hourly hydrogen demand for 6 ferries in  Norway.

H2_bus_demand

Synthetic hourly hydrogen demand based on daily bus operations (15 buses on weekdays, fewer on weekends).

Elz_storage

Specifications for two electrolyzers (1.2 MW and 2.5 MW) and associated hydrogen storage.

Grid_limits

Maximum power sell to Grid & Maximum power buy from Grid

Each component is provided in .xlsx format, organized by timestep. The dataset structure is designed for ease of integration into simulation tools and optimization models. The data can be read in various programming languages e.g. JuMp, Python, MATLAB, R etc. The units of all parameters are specified in the source file. The dataset can be accessed on zenodo platform.

3. Data generation methodology

PV Generation (Gen_pv)

The base data for 25 PV prosumers was used to simulate a wider adoption scenario. To reflect varied system sizes and orientation, synthetic profiles for an additional 275 users were created by scaling the original 25 profiles using a random multiplier between 0.8 and 1.2. This approach helps represent a future-oriented community with widespread rooftop solar adoption.

Wind Generation (Gen_wind)

Hourly wind generation data was obtained from renewables.ninja using a modeled Nordex N60 1.3 MW turbine. The selected site (Porsgrunn) and turbine parameters ensure that the generation profile reflects local weather conditions.

Load Data (Load)

Smart meter data from 400 real users was used without modification. The data includes a mixture of residential, commercial, and industrial demand. Visual inspection suggests the presence of EV charging in several household profiles, contributing to higher evening peaks.

Battery Data (Bat)

Since no battery data was included in the original dataset, battery systems were synthetically assigned to 250 users. Eight commercially available battery models were selected and distributed randomly among users. Each model includes specifications such as capacity, charge/discharge rate and efficiency.

Hydrogen Ferry Demand (H2_ferry_demand)

The demand data for six hydrogen ferries was generated based on earlier inputs, assuming fixed routes and consistent daily usage patterns typical of Norwegian ferry operations. 

Hydrogen Bus Demand (H2_bus_demand)

Synthetic hydrogen bus demand was created based on realistic bus specifications, operational conditions & schedules. A logical program was developed using Julia Mathematical Programming with following considerations.

·       15 buses run on weekdays, 10 on Saturdays, and 7 on Sundays.

·       Buses are assigned to 3 fixed routes (30, 36 & 42 km roundtrip distance).

·       Hydrogen consumption rate: 7.0 kg H₂ per 100 km.

·       Daily refueling occurs in one of three time windows (early morning, late morning, or afternoon).

·       Cold-weather adjustments (+7% consumption in Jan–Mar) and occasional shutdowns add realism. The national constitution day of Norway (17 May) is considered as holiday and hence no departures are scheduled.

·       One bus is selected and undergoes for a full day for maintenance every month.

Electricity Prices (Buy_price, Sell_price)

Buy prices are based on real 2024 hourly market data. Sell prices are estimated from Norwegian electricity statistics (Q3 2024), adjusted to exclude taxes and grid fees.

Maximum Power Import & Export (Grid_limits)

For the variety of end-users in the low-medium voltage system, the maximum contracted power exchange limits are specified as 10 kW, 15 kW, 25 kW for residential and commercial buildings. For industrial plant, the contracted power limit is kept 50 kW. These limits are assumed to keep the power import and export within specified limit.