Why Korea needs an open, realistic power grid test system and how KPG193 enables decarbonization research.

A practical explanation of why KPG193 matters for Korea’s energy transition.

Figure 1. Grid data is not publicly available!
Korea faces a unique challenge:
renewable penetration is still low, electricity demand is highly concentrated in Seoul,
and detailed grid data is not publicly available.
At the same time, researchers, policymakers, and planners need
realistic grid models to study decarbonization, renewable integration, grid congestion, and investment planning.
But until recently, Korea had only two options:
A realistic system with thousands of buses (not publicly available)
A simple 11-zone model (too coarse for serious analysis)
To fill this gap, the KPG 193 synthetic grid was created —
a publicly available, realistic, and computationally manageable model
that captures the key characteristics of the Korean power system.
Most transmission line locations, capacities, and equipment details are not publicly released.
To evaluate offshore wind, large-scale solar, hydrogen, storage, and interconnection options,
researchers need a realistic network model.
Huge demand concentration in the capital region
Major generation (coal, nuclear) located along the east coast
Solar mostly in the southwest
An isolated grid (no synchronous links to neighbors)
The KPG193 system mimics these core characteristics.

Figure 2. Data workflow used to build the synthetic Korean power grid.
KPG193 was created by combining multiple open datasets, each with different spatial resolutions:
OpenStreetMap (OSM) for approximate transmission line locations
KPX & KEPCO for power plant capacity, municipal electricity demand, and utility office boundaries
LDAPS weather data for solar and wind resource profiles
Population statistics to allocate demand realistically
Because exact grid data is unavailable, the authors used a clever clustering technique:
KEPCO’s 193 regional offices → 193 buses in the synthetic grid
This ensures the system preserves realistic geography
(demand clusters, coastal generation, renewable-rich regions)
without revealing sensitive information.
The final system includes:
193 buses
122 generators
407 transmission lines
HVDC link (Bukdangjin–Godeok, manually added)
Hourly demand & renewable profiles (2022)
MATPOWER-formatted network data
Fuel | Actual | KPG193 |
|---|---|---|
Coal | 32.5% | 33.9% |
Gas | 27.5% | 27.8% |
Nuclear | 29.6% | 28.2% |
Renewables | 9.6% | 10.1% |
KPG193 closely matches the national mix — a key requirement for realistic decarbonization studies.

Figure 3. Network topology and generator locations in KPG193.
KPG193 reflects the Korean grid’s core features:
(Shown in the upper circled region in Fig. 3 of the paper.)
Power flows predominantly northwest toward the capital.
Leads to overvoltage issues in weak-grid areas, also observed in real operation.
A major challenge for Korea’s renewable expansion.
These patterns appear clearly in KPG193’s ACOPF validation.
The model was tested using:
Daily Unit Commitment (UC)
Hourly AC-OPF (ACOPF) for all of 2022
Results show:
This means KPG193 is not just a synthetic dataset —
it faithfully reproduces operational characteristics of the real Korean grid.
KPG193 is built entirely from open data, so it cannot include:
Reactive compensation devices (SVC, STATCOM, shunt caps)
Exact transformer tap settings
Detailed substation layouts
Fully accurate OSM topology (OSM may be outdated)
Still, it serves as the most realistic publicly available Korean transmission test system.
Researchers, utilities, and policymakers can now use KPG193 to study:
KPG193 enables Korea to join global open-grid modeling efforts,
like SciGRID (EU), PyPSA-Eur, and the Texas backbone systems.
Song, G., & Kim, J. “KPG 193: A Synthetic Korean Power Grid Test System for Decarbonization Studies.” 2025 IEEE Power & Energy Society General Meetings (PESGM) [link]