article · IET Renewable Power Generation
Maintaining power grid balance requires electricity generation to match consumer demand, making accurate predictions of solar power output vital for network stability. Weather data and generation metrics were gathered across four seasons over a single year from an operational 20 kW grid-connected photovoltaic system. These onsite measurements served to evaluate several machine learning approaches, specifically random forest algorithms alongside deep neural networks, artificial neural networks, and support vector regression. When tested for short-term multi-step ahead projections, the random forest model delivered the strongest results. Specifically, it achieved performance gains of 49 percent for 15-minute horizons and 50 percent for 30-minute horizons compared to the baseline support vector regression model. This approach demonstrates a viable method for enhancing short-term solar power forecasting using seasonal on-site operational data.
Solar power output varies with weather and seasons, complicating efforts to keep electrical grids stable. Improving short-term forecast accuracy helps grid operators balance generation with electricity demand in real time. Better predictions reduce the operational uncertainty associated with integrating large-scale solar arrays into existing power networks, supporting reliable delivery of clean electricity to consumers.
The findings could be applied by electrical grid operators and solar farm managers seeking predictive software for short-term energy scheduling. The work represents applied and tested research, demonstrated on an operational 20 kW array using 15- and 30-minute horizons. Moving towards commercial software tools or grid management platforms would require further validation across diverse geographies, larger plant capacities, and integration into industrial dispatch systems.
AI-generated from the published abstract. Always read the original work before citing.
Abstract To maintain grid stability, the energy levels produced by sources within the network must be equal to the energy consumed by customers. In current times, achieving energy balance mainly involves regulating the electrical energy sources, as consumption is typically beyond the control of grid operators. For improving the stability of the grid, accurate forecasting of photovoltaic power output from largely integrated solar photovoltaic plant connected to grid is required. In the present study, to improve the forecasting accuracy of the forecasting models, onsite measurements of the weather parameters and the photovoltaic power output from the 20 kW on‐grid were collected for a typical year which covers all four seasons and evaluated the random forest techniques and other techniques like deep neural networks, artificial neural networks and support vector regression (reference in this study). The simulation results show that the proposed random forest technique for the forecasting horizon of 15 and 30 min is performing well with 49% and 50% improvements in the accuracy respectively over reference model for the study location 22.78°N, 73.65°E, College of Agricultural Engineering and Technology, Anand Agricultural University, Godhra, India.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1049/rpg2.12921
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
