Improvement of Methods for Predicting the Generation Capacity of Solar Power Plants: the Case of the Power Systems in the Republic of Crimea and City of Sevastopol

Abstract

The construction and operation of large solar power plants (SPPs) and the dependence of their production on light and other meteorological factors leads to a strong dependence of the operation modes of the Republic of Crimea and Sevastopol power system on meteorological factors. Today, given that the share of solar power plants is about 30% of the total installed capacity, it is necessary to solve the problems that have a great impact on the power system operating modes. With large output capacity of the solar power plant, the operator has to give commands to turn off the generating equipment of thermal power plants. In power systems with a large share of solar generation, it is necessary to solve this problem by improving the generated power predicting methods, as it will reduce the dependence of operating modes on weather factors and increase the reliability of the power system. The paper discusses the use of hybrid predicting methods that imply taking into account the possibility of the weather scenarios simulation, advanced cloud-based image processing technology, and close-to-real-time cloud motion surveillance cameras. There was an experimental software created that selects coefficients of set configuration time series. In combination with the conservative methods, it makes predicting the SPP Perovo output more accurate. Taken together, the chosen methods of predicting solar power generation capacity in the power system of the Republic of Crimea and Sevastopol ensure not only stability of the power system as a whole, but also the maximum efficiency of power plants, allow to accelerate the integration of solar power plants into the power system, and have positive effects on the environment.