Обзор и классификация методов оценки и прогнозирования солнечной радиации

Abstract

Various methods used to estimate and predict solar radiation are reviewed. Ground-based solar radiation intensity measurements are not carried out everywhere, despite the fact that the need to estimate the gross (theoretical) potential of solar energy in a region arises quite often. In such cases, approximate estimates of solar energy resources are carried out using various models, the main ones of which are considered in this paper. The availability of an adequately estimated local solar energy potential is important not only for rational distribution of funds in designing, but also for predicting the output of a power plant that uses solar energy. A general classification of methods for estimating and predicting solar radiation by the type of used parameters and estimated quantities (physical and statistical models) is presented. Statistical methods include conventional empirical models, in which the monthly average daily sums of solar radiation on a horizontal surface are estimated with dividing them into 4 categories depending on the input parameters involved, and models that evaluate the sums of the solar radiation diffuse component on horizontal (decomposition models and parametric models) and inclined (isotropic and anisotropic models) receiving areas. Models based on data mining methods are also referred to statistical methods for estimating and predicting solar radiation. The suggested statistical models are based on various parameters such as sunshine duration, maximum ambient temperature, relative humidity, pressure, latitude and longitude of the locality, the day of year ordinal number, cloudiness index, average ambient temperature, altitude, average wind velocity, precipitation, etc. The results of modeling the monthly average daily amounts of total solar radiation on a horizontal surface according to the universal empirical models are evaluated and compared with the long-term average data of ground-based measurements on the territory of Kyrgyzstan and with the data from the NASA POWER database.