ИСТИНА |
Войти в систему Регистрация |
|
Интеллектуальная Система Тематического Исследования НАукометрических данных |
||
A method has been developed for refining a short-term forecast of solar power plant performance using machine learning, which allows reducing financial losses of inaccurate forecast in most cases to 2-7% of the generated electricity cost and is approximately an order of magnitude more accurate than the dynamic simulation of solar power plant operation based on numerical weather prediction data.