Modeling and Improving the Accuracy of Solar PV Output Forecasts

The large-scale integration of solar energy into national and regional power systems presents notable challenges within the global transition to renewable energy. As the deployment of solar generation increases, particular attention must be paid to the forecasting accuracy and its impact on power system stability with grid-connected solar power plants (SPPs) or solar farms.

The variability and intermittency of solar energy, driven by stochastic meteorological conditions, make the reliable operation of power systems increasingly dependent on precise generation forecasts. Accurate forecasting of electricity production from solar farms is essential for efficient system operation, balancing, planning, and market activities. Forecast accuracy is influenced by weather conditions, seasonal fluctuations, and the technical characteristics of generation and monitoring equipment.

Enhancing the precision of solar generation forecasting plays a critical role in facilitating the stable and efficient operation of power systems. Important factors influencing solar generation include: solar elevation and azimuth angles, air temperature, cloud coverage, surface condition, and the presence of fog or precipitation. These factors must be considered in predictive models.

Additionally, incorporating real-time data on equipment condition and field performance of solar farms can significantly improve forecast reliability and power system responsiveness. Key application areas include:
– Power system operations and management
– Electricity market and trading strategies
– Operation planning of grid-connected solar farms
– Integration with battery energy storage systems (BESS)
– Development and optimization of virtual power plant (VPP) architectures

Research submitted to this Research Topic should aim to improve forecasting methodologies and adapt them to modern energy system requirements, particularly through integration with smart grid technologies and control systems. Emphasis should be placed on advanced approaches such as machine learning, deep learning, hybrid AI-based techniques, probabilistic forecasting, and the use of satellite and meteorological data.

We welcome contributions based on both theoretical advancements and practical case studies supported by simulation results or operational data from real-world solar installations. Topics of particular interest include, but are not limited to:

- Forecasting of solar generation using weather and satellite data.
- Integration of solar forecasting tools into smart grid energy management systems (EMS).
- Probabilistic and hybrid forecasting models.
- Demand response and dispatch optimization.
- Storage system integration and control.
- Evaluation of forecasting errors and their impact on power system reliability.
- Forecasting within microgrids and decentralized systems.
- Virtual power plant operation strategies.
- Power balancing and reserve capacity planning for forecast deviation compensation.