Summary
The paper proposes a process for system level flexibility assessment using electricity market modeling, load-flow estimation together with transfer capability estimation using sensitivity analyses, and machine learning regression models. The paper demonstrates the applicability of the proposed process for a selected transmission corridor in the Finnish 400 kV transmission system and estimates the anticipated flexibility needs in 2026 using three distinct weather scenarios.
Read more Read lessIn Finland, the changes in the generation and load types have increased the volatility of the network power flows and operational conditions. Finnish transmission network is operated with
N−1 criterion during both normal and outage situations. Therefore, understanding the possible variations in the network power flows and transfer capabilities are crucial for system operations.
This enables the transmission system operator (TSO) to securely maximize the transfer capacity given to the electricity market, and prepare for different operational situations. Thus, the flexibility assessment method must consider the variations of the transmission network flows and transfer capabilities of the system, and their relations with varying climate conditions. The proposed flexibility assessment process utilizes modeling of the day-ahead electricity market and hourly time-series AC load-flow estimates under the studied weather scenarios. The transfer capability of the transmission network is calculated using AC load-flow for a combination of selected operational scenarios. The selection is based on the sensitivity of the transfer capability to observable operational parameters, such as system load and wind production. Machine learning regression models are used to determine the transfer capability for each hour of the market model results.
A key benefit of the regression models is that they enable estimation of the transfer capability of the system under uncertain climate conditions without time-consuming deterministic loadflow simulations for each hour. This enables the statistical analysis of the transmission network bottlenecks and flexibility needs.
The paper presents hourly level statistical analyses of flexibility needs in a selected transmission corridor in the Finnish 400 kV transmission network in 2026. The study is performed using three distinct weather scenarios for the years 1999, 2000, and 2010.
The results show that in the studied transmission corridor, the flexibility is typically needed during cold winter season. The results show that total system demand, electric heating, transit flow, and wind power production are significantly higher when flexibility is needed, compared to situations when flexibility is not needed. Also, the results show that flexibility durations per incident vary from 1 to 17 hours, typically within the range of 1–5 hours.
The study presented in the paper focuses on a situation where transfer capability of the transmission system is limited by either thermal or voltage limits. As discussed in the paper, the proposed flexibility assessment method could also be applied to a case where other phenomena, such as electromechanical oscillations or converter driven instability, limit the capability. However, further studies regarding these situations are needed.
Additional informations
| Publication type | Session Materials |
|---|---|
| Reference | C2_11373_2026 |
| Publication year | |
| Publisher | CIGRE |
| Country | Finland |
| Study committees | |
| File size | 436 KB |
| Price for non member | 30 € |
| Price for member | 30 € |
Authors
NIKKILÄ Antti-Juhani - Fingrid Oyj; KANERVA Markus - Fingrid Oyj; VIRTANEN Markus - Fingrid Oyj; VÄISÄNEN Leevi-Kalle - Fingrid Oyj; IHALAINEN Erkko - Fingrid Oyj; ASP Tommi - Fingrid Oyj
Keywords
Flexibility, Machine Learning, Operational Planning, Power System Operation, Time-Series Load Flow