Abstract
Policies promoting a carbon-neutral economy are accelerating the energy transition towards renewable electricity generation, primarily through small-scale investments within distribution networks. As penetration of distributed generation increases, transmission system operators encounter increasing challenges in predictability. This study presents a coordination approach between transmission and distribution network operators in Türkiye, highlighting an advanced solar generation forecasting method to enhance secure network operations. In the initial phase, a centralized database was established to catalog distributed generation capacity and plant specifications. Web-based applications and services were developed to integrate distribution company assets into the transmission network model. By September 2024, over 30,000 power plants (99% photovoltaic, 17.8 GW total) had been integrated. Distribution companies are expected to supply real-time generation data to the transmission operator, though issues such as measurement errors and communication problems may limit data accuracy and continuity. Thereafter, a short-term solar forecasting system was developed to support intraday and day-ahead congestion analysis. The system provides high-resolution forecasts at the individual plant level for more than 20,000 locations. Machine learning algorithms were used for large-scale plants, while small-scale plant forecasts were based on meteorological data and physical modeling. Because accurate forecasting requires detailed plant parameters, a Particle Swarm Optimization algorithm was implemented to calibrate photovoltaic models. This approach enables accurate parameter estimation even when on-site data is limited, improving the performance and reliability of solar generation forecasts. This paper provides valuable insights into the coordination between transmission system operators and distribution system operators, emphasizing the importance of data integration and forecasting accuracy in managing distributed solar generation. Furthermore, it details the outcomes of an advanced solar generation forecasting system, demonstrating its effectiveness in improving predictability and supporting secure and efficient power system operations.