Scientific Journal of KubSAU

Energy complex of the country is a collection of electrical installations high and low voltage, producing, transforming, transporting, distributing and consuming electricity. A huge web of networks and more than 700 generating plants with a total capacity of about 230 GW [1]. Almost 90% of this potential is formed in a unified technical complex of the Unified energy system (UES) [2]. Now there is a deterioration of the performance of the sector. In comparison with 90-mi years of XX century, more than 1.5 times increased power losses in the power grids. Significantly increased the proportion of obsolete electrical and auxiliary equipment, and the load on the network every day only increase. All speaks of the necessity of both local and global modernization of the electric power complex of the country and the creation of a new concept of consumption management and energy distribution in the network. In the framework of the energy strategy of the Government of the Russian Federation dated 13 November 2009 # 1715-R, to improve handling and ensure reliable operation of electric power systems, wider introduction of flexible transmission system (FACTS devices) and improvement of systems of automatic emergency protection and dispatching control [1]. The development of electric power complex of the country should be in the way of intelligent networks. This is possible through the use of modern components that can make the process of managing "intelligent". In foreign literature, this term is called the Smart Grid

The growth of breakdown in electric networks of Russia has been followed by the growth of industrial and domestic power consumption taking place in our country in the last decade. This situation is directly connected with the high percentage of outdated equipment in electric networks. Such thing as a loss of power is directly connected with deterioration of equipment and breakdown. Average losses of power in the country are at the level of 1980-s, despite the turning point to the side of reduction outlined recently. Forecasting is one of the activities that contribute to their reduction, as it allows identifying adverse trends and calculating the effect of different kinds of technological arrangements. The visible growth of interest in the intellectualization of education and in the different fields of industry should not avoid the electric power field, which is quite conservative. The development of the concept of a “smart network” and its implementation would help to improve the reliability of electric networks. One of the signs of such a network is the possibility to assess the current situation automatically and forecasting of its parameters in the future, including energy losses. This article examines the main factors affecting the value of power losses, the analysis of the most popular methods of forecasting is conducted, and conclusion about the prospects of their use to predict the losses of power has been made based on the results of this analysis

Nowadays the high level of electricity losses is one of the most important issues of the energy industry in the Russian Federation recognized at the state level. According to many sources, one of the activities that contribute to reduce energy losses, is their planning, an important component of which is the prediction of electricity losses on the basis of retrospective information. The highest percentage of technical losses of electricity is accounted for distribution network with a voltage range 0,4-35 kV. In this regard, the most productive activity is forecast construction namely of this component of power losses. According to some features of the regarded value (electricity losses) the most effective activity for its forecasting is using methods with artificial intelligence elements. One of these methods, having a number of important advantages, is forecasting fuzzy time series. This technique is widely consecrated in foreign publications, but did not find sufficient popularity in our country. This article analyzes the existing models of forecasting fuzzy time series on the basis of which proposals for their improvement and adaptation in order to predict the loss of electricity are made; designed model of multivariate fuzzy time series forecasting of energy losses is given