Search academic texts

Information × Registration Number 0226U001215, (0125U002908) , R & D reports Title Development of mathematical and software-information tools for forecasting energy generation from solar installations for integration into building energy supply systems popup.stage_title Аналіз підходів до прогнозування виробництва енергії сонячними установками та розроблення математичної моделі короткострокового прогнозування виробництва електричної та теплової енергії із застосуванням технологій машинного навчання Head Karpenko Dmytro S., Кандидат технічних наук Registration Date 22-01-2026 Organization Institute of General Energy National Academy of Sciences of Ukraine popup.description1 Development of mathematical and software-information tools for short-term forecasting of energy volumes to ensure thermal comfort and electrical energy from solar installations integrated into building energy supply systems using machine learning technologies. popup.description2 Object of research – the processes of electrical and thermal energy generation by various types of solar installations integrated into building energy supply systems. Purpose of the study – the development and mathematical formalization of a hybrid short-term forecasting model for electrical and thermal energy production by solar installations integrated into building energy supply systems, using machine learning methods, based on the analysis of solar installation types, available datasets, and modern forecasting approaches. A comprehensive analytical review of solar installation types, contemporary forecasting approaches, and datasets covering different time scales and operating regimes of photovoltaic, thermal, and hybrid PV/T systems was conducted. Scientific and technical limitations in existing approaches were identified, including the lack of universal models for solar systems, insufficient long-term measurements considering degradation, panel soiling, and climate change, as well as the insufficient development of methods for joint forecasting of electrical and thermal power. A mathematically formalized hybrid forecasting model was developed, including the definition of input data structure, time series modeling, integration of deep learning components, and an ensemble scheme for their combination. The study employed: systems analysis; comparative analysis; and the use of computational techniques. Product Description popup.authors Dmytro S. Karpenko popup.nrat_date 2026-01-22 Close