The paper addresses the problem of short-term renewable energy forecasting. The stochastic nature of renewable energy sources affects the power system planning procedures, subsequently reducing the reliability and security of power supply for final customers. The authors propose a machine learning approach for effective day-ahead forecasting using retrospective metering data and open source weather information provided by meteorological services. The paper addresses the problem of feature identification and appropriate error metrics. For the purpose of effective testing the user interface was created and tested on the real solar power plant situated in southern region of Russian Federation. The mean forecasting accuracy was obtained to be about 93%.

This article highlights the industry experience of the development and practical implementation of a short-term photovoltaic forecasting system based on machine learning methods for a real industry-scale photovoltaic power plant implemented in a Russian power system using remote data acquisition. One of the goals of the study is to improve photovoltaic power plants generation forecasting accuracy based on open-source meteorological data, which is provided in regular weather forecasts. In order to improve the robustness of the system in terms of the forecasting accuracy, we apply newly derived feature introduction, a factor obtained as a result of feature engineering procedure, characterizing the relationship between photovoltaic power plant energy production and solar irradiation on a horizontal surface, thus taking into account the impacts of atmospheric and electrical nature. The article scrutinizes the application of different machine learning algorithms, including Random Forest regressor, Gradient Boosting Regressor, Linear Regression and Decision Trees regression, to the remotely obtained data. As a result of the application of the aforementioned approaches together with hyperparameters, tuning and pipelining of the algorithms, the optimal structure, parameters and the application sphere of different regressors were identified for various testing samples. The mathematical model developed within the framework of the study gave us the opportunity to provide robust photovoltaic energy forecasting results with mean accuracy over 92% for mostly-sunny sample days and over 83% for mostly cloudy days with different types of precipitation.

The paper describes the methodology for consideration of renewable energy sources in strategic plans of fuel and energy sector development. The inclusion of renewable energy projects in energy development programs is carried out based on the results of an integrated ranking of technical, economic and environmental criteria. For power generation facilities based on renewable energy sources, the energy potential is additionally addressed in accordance with the list of power generation technologies under consideration, which makes it possible to obtain the values of the installed capacity utilization factors, and, as a consequence, to judge on the technological and commercial feasibility of renewable energy projects implementation. The paper provides an example of a ranked list formulation of renewable energy projects for a real regional power system.

The paper discusses a practical case of creating an autonomous microgrid for remote residential customers power supply. Due to the specifics of the geographic location, the main scenarious, considered within the framework of the study are : distribution network conncetion, creation of a diesel power plant, creation of an autonomous hybrid micro-grid based on wind power plants and energy storage. In the study, a load forecast has been made and a methodology for determining the installed capacity of the main generation sources is presented. According to the criteria of technological efficiency and environmental safety, the option of creating an autonomous microgrid was adopted for implementation.